The impact of profitability, firm size, and liquidity on firm value

with capital structure as a moderating variable: Evidence from the

manufacturing firms on HOSE
1. RESEARCH OBJECTIVES
1.1 General objective of study
The thesis aims to determine the impact of profitability, firm size, and
liquidity on firm value with capital structure as a moderating variable. From
that, the author can give implications to assist firms in making proper
decisions in managing the impact of these factors on firm value.
1.2 Specific objectives of study
Objective 1: Evaluating the current situation of the Ho Chi Minh Stock
Exchange and the firm value of manufacturing firms on HOSE.
Objective 2: Exploring the impact of profitability, firm size, and
liquidity on firm value with capital structure as a moderating variable from
manufacturing firms on HOSE.
Objective 3: Proposing managerial implications to improve the firm
value of manufacturing firms on HOSE.
 LITERATURE REVIEW
2.1 LITERATURE REVIEW
Several studies used signaling theory to show the factors that influence
firm value, Pratiwi (2020) argued that signaling theory demonstrates how
businesses can communicate with stakeholders by providing information in
various formats, which stakeholders can use to make informed decisions.
Therefore, the authors researched to examine the influence of capital structure,
profitability, and firm size on firm value using data from companies listed on
the Indonesian Stock Exchange. The results indicated that profitability had a
positive impact on firm value while capital structure had a negative impact on
firm value. Meanwhile, Ahmad & Muslim (2022) showed similar results when
using the Ordinary least square model and carried out in Indonesia. Similarly,
Margono & Gantino (2021) also relied on signaling theory and conducted
empirical research on the impact of business size, leverage, profitability, and
dividend policy on the firm value of the consumer goods industry in the food
and beverage sub-sector listed on IDX from 2016 to 2019. The results showed
that leverage, profitability, and dividend policy had a positive impact on firm
value, while firm size did not affect the value of the business. Furthermore,
Tarigan et al., (2022), examined data from manufacturing sector companies
listed on the Indonesia Stock Exchange from 2016 to 2020. The results of the
study showed that company size and investment decisions had a positive effect
on firm value while diversification strategies had a negative effect. Besides,
capital structure was able to moderate the relationship between company size,
diversification, investment decisions, and firm value.
The Agency theory was also considered in some previous studies,
Agency theory is considered a game theory that establishes a contractual
relationship between the agent and the principal (Hidayat et al., 2019). Based
on agency theory, Amarudin et al., (2019) investigated the effect of growth
opportunity, corporate tax, and profitability on firm value in the manufacturing
sector companies listed on the Indonesian Stock Exchange. The authors used
path analysis to analyze the data and found that growth opportunity and
corporate tax had no significant impact on firm value. However, the authors
discovered that capital structure and profitability positively influenced firm
value. Meanwhile, in addition to using agency theory, the research of Goh &
Rumapea (2019) also used a combination of signal theory. The authors used
multiple linear regression analysis to analyze the data and show the results that
profitability and company size had a positive influence on firm value, while
capital structure significantly affected firm value but with a negative
coefficient direction. Furthermore, Husna (2020) also used two similar
theories and found that managerial ownership and firm size had a positive
impact on firm value while capital structure had a negative impact on firm
value. Not only that, the author also proved that capital structure can moderate
the impact of managerial ownership and firm size on firm value.
Furthermore, some studies in Vietnam also used agency theory to study
the impact of factors on firm value, Tran (2019) applied various theories
including Modigliani and Miller’s theory, agency theory, tradeoff theory,
pecking order theory, and market timing theory. In this study, the relationship
between capital structure and firm value was examined using a quantile
regression method based on panel data from 3122 observations. The results of
the quantile regression analysis showed a positive impact of capital structure
on firm value. Additionally, the quantile regression approach provided insights
into the relationship between firm financial characteristics (such as firm size,
asset tangibility, growth opportunities, dividend payout, liquidity, and state
ownership) and firm value. Moreover, Nguyen & Nguyen (2017) utilized
agency theory to investigate the correlation between stock liquidity, corporate
governance, and firm value. The authors conducted their study using a dataset
of 635 non-financial companies listed on HSX and HNX between 2007 and
2015. The authors employed seemingly unrelated regression (SUR) to test
hypotheses. The findings indicated a positive relationship between stock
liquidity and corporate governance, as well as a positive impact of corporate
governance on firm value. Based on agency theory, the issue of private profits
of major shareholders in developed countries like Vietnam is becoming
serious, Vo (2014) examined the relationship between ownership structure and
firm value on the Ho Chi Minh City Stock Exchange. The results of the study
reveal that foreign shareholders and institutional shareholders had a positive
impact on firm value.
The previous models relate to firm value
Additionally, based on signaling theory, a study by Fajaria & Isnalita
(2018) examined the impact of profitability, liquidity, leverage, and firm
growth on firm value, with dividend policy as a moderating variable. The
study applied both signal theory and contingency theory, using four
independent variables: profitability, liquidity, leverage, and firm growth. The
moderating variable was dividend policy, and the control variable used in the
model was firm size. The study collected 396 observations from the official
website IDX, and the data were analyzed using regression analysis moderation
(MRA). The research results revealed that profitability and growth had a
positive influence on firm value. On the other hand, liquidity and leverage had
a negative impact on firm value. Furthermore, the study found that dividend
policy strengthened the liquidity effect on firm value. Conversely, dividend
policy weakened the influence of profitability and leverage on firm value.
However, dividend policy was not able to moderate the effect of growth on
firm value.
PROFITABILITY H1

LIQUIDITY H2
H6
H7 FIRM VALUE
H3
LEVERAGE

H4
FIRM GROWTH

H5 H8

DIVIDEND POLICY FIRM SIZE

Figure 2.1: Research model by Fajaria & Isnalita, 2018

Nguyen et al., (2021) researched the determinants of firm value and
empirical research on listed trading companies in Vietnam. The authors
proposed and tested a model with five independent variables: firm size, capital
structure, firm sales, profitability, and liquidity. The study utilized 925
observations from 239 trading companies listed on the Vietnamese stock
market, with firm value measured by the book value of equity. The collected
observations were analyzed using descriptive statistics, including measures
such as the minimum, maximum, average, and standard deviation. The study
employed fixed effects models and random effects models, and the Hausman
test was used to select the appropriate model. The findings of the study
revealed that capital structure had a negative impact on firm value. However,
the variables of profitability, firm sales, and liquidity were not statistically
significant in the regression model, indicating that these factors did not affect
firm value significantly. Additionally, the study highlighted that firm size was
the most important factor determining firm value. It is important to note that
this study has a limitation as it focuses solely on listed trading companies, thus
only identifying the factors that affect the value of listed trading companies in
Vietnam. Furthermore, the dataset used in the study spans from 2011 to 2019,
and therefore, the data is not updated beyond 2020.

Firm size

Capital structure

Firm sales Firm value

Profitability

Liquidity

Figure 2.2: Research model by Nguyen et al., 2021

In addition, Dang et al., (2019) analyzed the impact of growth, firm size,
capital structure, and profitability on enterprise value. This study used 214
non-financial corporations listed on the Ho Chi Minh Stock Exchange, with
sufficient information for a 5-year analysis, resulting in a total of 1070
samples. The research not only examined the independent variables that affect
enterprise value but also considered the impact of growth, firm size, and
capital structure on profitability. Enterprise value was measured by Tobin's Q
and EV were used. The research employed quantitative research methods,
utilizing regression models with OLS, FEM, and REM methods. In cases
where the regression model exhibited heteroskedasticity and autocorrelation,
the GLS method was employed to address these defects. The study revealed
that capital structure has a negative effect on enterprise value when measuring
it using Tobin's Q or EV. Furthermore, when measuring enterprise value using
Tobin's Q and EV, firm size was found to have a positive influence on
enterprise value. Additionally, profitability (ROE) showed a positive and
statistically significant correlation with enterprise value when measured by
EV, while profitability (ROA) had a positive impact on enterprise value as
measured by Tobin's Q.
 Capital structure

H3
H2b
H3a

H2 Enterprise
Size Value
EV, Tobin’s Q
H2a
H4

Profitability
(ROA/ROE)
H1
H2c
H1a

Growth

Figure 2.3: Research model by Dang et al., 2019

Research conducted by Indira & Wany (2021) examined how capital
structure moderates the impact of liquidity and profitability on firm value.
This study specifically focuses on the context of Indonesia, utilizing data
collected from 96 companies spanning the years 2016 to 2019. To investigate
the influence of liquidity and profitability on firm value, regression analysis
was used, with capital structure as a moderating variable. In addition, the
study utilized moderated regression analysis (MRA), a regression test that
incorporates a moderator variable in the regression equation by including an
interaction term, specifically the multiplication between the moderator
variables. The results of this study have proven the results if liquidity has a
negative and significant impact on the firm value, while profitability has a
positive impact on the firm value. Moreover, capital structure weakens the
negative impact of liquidity on firm value but does not moderate the influence
of profitability on firm value. Additionally, the study suggests that expanding
the scope of the research to include other industries such as manufacturing,
property, and mining companies would enable comparisons to be made and
contribute to a broader understanding of the research findings.
 (X1)
LIQUIDITY
(LDR)

(Y)
FIRM VALUE
(PBV)

(X2)
PROFITABILITY
(ROA)

(Z)
CAPITAL STRUCTURE
(DER)

Figure 2.4: Research model by Indira & Wany, 2021

Table 1.1: Summary of literature review related to the research topic

No Research Title Researcher Samples Theory Research Variables Research result

Methods
1 The effects of Tarigan et 188 - Signaling - Fixed Dependent variable - Company size positively
company size, al., (2022) Theory effect - Firm value influences firm value.
business model (measured by PBV) - Diversification strategy
diversification, (FEM) Independent variables negatively influences firm
and investment - Company Size value.
decision on firm - Diversification - Investment decisions
value with capital - Investment decision positively influences firm
tructure as a Moderating variable value.
moderating - Capital Structure - Capital structure is able to
moderate the relationship
between company size,
diversification, investment
decision, and firm value.
2 The effect of Husna 66 - Signaling - The Dependent variable - Managerial ownership
managerial (2020) Theory multiple - Firm value variable has a positive impact
ownership and - Agency regression (measured by on firm value.
company size on Theory analysis Tobin’s Q) - Company size has a positive
firm value with (Principal- Independent variables impacts on firm value.
capital structure Agency - Managerial - Capital structure has a
as moderating Theory) Ownership negative impact on firm
variables in - Company Size value.
manufacturing Moderating variable - Capital structure is able to
companies listed - Capital Structure moderate the relationship
in Indonesia between managerial
Stock Exchange ownership and firm value.
 2016-2018 - Capital structure is able to
moderate the relationship
between company size and
firm value.
3 Do capital Pratiwi 155 - Signaling - Multiple Dependent variable - Capital structure has
structure, (2020) Theory Linear - Firm value anegative impact on firm
profitability, Regression Independent variables value.
and firm size Analysis - Capital structure - Profitability has a positive
affect firm value? - Profitability impact on firm value.
- Firm size - Firm size has no impact on
firm value.
4 Several factors Ahmad & 48 - Agency - Using Dependent variable - Total asset turnover has a
affecting firm Muslim theory multiple - Firm value negative impact on firm
value (2022) - Signaling linear (measured by PBV) value.
manufacturing in theory regression Independent variables - Debt to equity ratio has a
indonesia analysis - Total asset turnover negative impact on firm
with the - Debt to equity ratio value.
Ordinary - Current ratio - Current ratio has a positive
Least - Return on equity impact on firm value.
Square - Return on equity has a
model positive impact on firm value.

5 The influence of Margonno 40 - Signaling - Multiple Dependent variable - Firm size, leverage,
firm size, & Gantino Theory linear - Firm value profitability, and dividend
leverage, (2021) regression (measured by PBV) policy simultaneously
profitability, and analysis Independent variables influence firm value.
dividend policy - Company size - Firm size has no impact on
on firm value of - Leverage firm value.
 companies in - Profitability - Leverage has a positive
Indonesia stock - Dividend policy impact on firm value.
exchange - Profitability has a positive
impact on firm value.
- Dividend policy has a positive
impact on firm value.
6 The improve level Hidayat et 635 - Agency - Multiple Dependent variable - Liquidity has a positive
of firm value with al., (2019) theory linear - Firm value influence on firm value.
liquidity, debt regression Independent variables - Debt policy has no impact on
policy and analysis - Liquidity firm value.
investment in - Debt policy - Investment decisions have a
indonesian - Investment positive impact on the firm
emerging market value.
7 The impact of Tran (2019) 3122 - Modigliani - OLS Dependent variable - Capital structure has a
capital structure and Miller - FEM - Firm value negative effect on firm value.
on firm value of theory - Quantile Independent variables - Firm size has a positive effect
Vietnamese listed regression - Capital structure
- Agency on firm value.
companies – a Control variables
quantile theory - Firm size
- Asset tangibility has a

regression - Tradeoff - Asset tangibility negative impact on firm

approach theory - Growth value.
- pecking order opportunites - State ownership has positive
theory - Dividend payout effect on firm value.
- Market - Liquidity
timing theory - State ownership
8 Effect of growth Amarudin et 160 - Agency - Path Dependent variable - Growth opportunity has no
opportunity, al., 2019 Theory analysis - Firm value impact on firm value.
corporate tax, and - Pecking (measured by PBV) - Corporate tax has no impact
profitability Order Theory on firm value.
 toward value of Independent variables - Profitability has a positive
firm through - Growth opportunity impact on firm value.
capital structure - Corporate tax
(Listed - Profitability
manufacturing
companies of
Indonesia)
9 The influence of Goh & 48 - Agency - Multiple Dependent variable - Capital structure has a
capital structure, Rumapea Theory linear Firm value negative effect on firm value.
profitability, and (2019) - Signal theory regression Independent variables - Profitability has a positive
company size on analysis - Capital Structure effect on firm value.
firm values in - Company Size - Firm size has a positive effect
manufacturing - Profitabilities on firm value.
companies listed - Capital structure, profitability
on the Indonesia and firm size simultaneously
Stock Exchange have a significant effect on
firm value.
10 The effect of Fajaria & 396 - Signal theory - Multiple Dependent variable - Profitability has a positive
profitability, Isnalita - Contingency Linear - Firm value impact on firm value.
liquidity, leverage (2018) theory Regression Independent variables - Liquidity has a negavtive
and firm growth Analysis - Profitability impact on firm value.
of firm value with - The - Liquidity - Leverage has a negative
its dividend moderated impact on firm value.
- Leverage
policy as a regression - Firm growth has a positive
- Firm growth
moderating analysis impact on firm value.
variable (MRA) Moderating variable - Dividend policy can
- Dividend policy strengthen the positive impact
of liquidity on firm value.
- Dividend policy can weaken
the positive impact of
 profitability on firm value.
- Dividend policy can weaken
the negative impact of
leverage on firm value.
11 Determinants of Nguyen et 925 - FEM Dependent variable - Capital structure has a
firm value: an al., (2021) - REM - Firm value negative impact on firm
empirical study Independent variables value.
of listed trading - Capital Structure - Firm size has a positive on
companies in - Profitability
Vietnam firm value.
- Revenue
- Liquidity
12 Study the impact Dang et al., 1070 - Agency - Compare Dependent variable - Growth has no impact on
of growth, firm (2019) Theory and select - Enterprise value enterprise value.
size, capital - Theory of among (EV) - Firm size has a positive
structure, and Information ordinary Independent variables impact on enterprise value.
profitability on Asymmetry least - Growth - Capital structure has a
enterprise value: Signaling square, - Firm size negative impact on enterprise
evidence of Theory fixed - Capital structure value.
enterprises in effects Profitability - Profitability has a positive
Vietnam model, and impact on enterprise value.
random
effects
model.
13 Capital structure Indira & 96 - Signalling - Multiple Dependent variable - Liquidity has a negative
as a moderation Wany theory Linear - Firm value impact on the firm value.
variable of the (2021) - Trade Off Regression Independent variables - Profitability has a positive
effect of liquidity Theory Analysis - Liquidity impact on the firm value.
and profitability - The - Profitability - Capital structure is able to
on firm value moderated
 (case study of regression Moderating variable moderate liquidity on firm
banking analysis - Capital structure value
companies on the (MRA) - Capital structure is not able to
idx 2016-2019)
moderate profitability on firm
value.
14 The relationship Nguyen & 4131 - Agency - Seemingly - Stock liquidity - Corporate governance
between liquidity, Nguyen theory unrelated - Corporate positively impacts firm value.
corporate (2017) regression governance - There is a positive causal
governance, and (SUR) - Firm value relationship between volume
firm valuation: and corporate governance.
Evidence from
Vietnam

15 Ownership Vo (2014) 1470 - Agency - FEM Dependent variable - Foreign ownership has a
structure and theory - REM - Firm value positive impact on firm value.
business value: - Dynamic Independent variables - Institutional ownerships has a
Research on GMM - Foreign ownership positive impact on firm value.
Vietnam stock - State ownership - State ownership has no
market
- Management impact on firm value.
ownership - Management ownership has
- Institutional no impact on firm value.
ownership
 After reviewing previous studies conducted on factors affecting firm
value and capital structure moderating the impact of factors on firm value, the
author found that there are several research gaps in this field.
Firstly, previous studies examining the impact of profitability, firm size,
and liquidity on firm value have not yielded consistent research results.
Pratiwi (2020) and Margonno & Gantino (2021) argued that profitability had a
positive impact on firm value. In contrast, Alisa & Aryani (2022) found that
profitability had a negative effect on firm value. Husna (2020) and Tarigan et
al., (2022) firm size had a positive effect on firm value. However, some
studies such as Pratiwi (2020) and Margonno & Gantino (2021) stated that
firm size did not affect firm value. Furthermore, while Hidayat et al., (2019)
argued that liquidity affects firm value, this result is not consistent with the
research results of Fajaria & Isnalita (2018) and Indira & Wany (2021).
Therefore, there is no solid basis to propose management implications to
enhance firm value.
Secondly, firm value was measured using various indicators that lack
uniformity, such as Tobin's Q, EV, PBV, and EPS. As a result, even within the
same research context, different measures can yield varying impact results.
Additionally, previous studies have also shown inconsistent findings regarding
the independent variables that affect firm value. Therefore, further exploration
of this gap is necessary.
As a result, to address the existing research gaps, this study aims to fill
them by investigating the impact of profitability, firm size, and liquidity on
firm value, using capital structure as a moderating variable. Firstly, Tobin's Q
index is utilized as a research measure. Tobin's Q ratio provides more
comprehensive information for expressing a company's value as it
incorporates all elements of the company's debt and share capital, including all
assets owned by the company. Secondly, research collects data on
manufacturing firms on the Ho Chi Minh Stock Exchange in the period 2018
to 2022, during this period the world economy in general and Vietnam's
economy have many fluctuations, which affects the financial situation of
manufacturing firms. Thirdly, given the inconsistent results in previous studies
regarding the impact of factors on firm value, this study aims to contribute to
the existing body of knowledge by examining the influence of profitability,
firm size, and liquidity on firm value, with capital structure as a moderating
variable. Capital structure can adjust other financial factors in the company.
Zulfa et al., (2022) also support this view, emphasizing the importance of
capital structure in shaping a company's financial position. Therefore, the new
point of the study is to examine how the capital structure will moderate the
effects of profitability, firm size, and liquidity on firm value in the period
2018–2022. This period corresponds to a time when Vietnam's economy
experienced significant fluctuations due to the impact of the COVID-19
epidemic and the global economic situation. Manufacturing firms, in
particular, were heavily affected in terms of production and financial
performance by these changes. Subsequently, the author will propose
management implications to enhance the value of manufacturing firms on Ho
Chi Minh Stock Exchange.
2.2 Research Model
To explain the impact of profitability, firm size, and liquidity on firm
value with capital structure as a moderating variable: Evidence from the
manufacturing firms on HOSE, signaling theory (Michael Spence, 1973), and
agency theory (Jensen & Meckling, 1976) is used as a scientific basis to form
a research model. The measurements used to calculate firm value are Tobin's
Q. Based on previous models, the research model for this study is proposed
below:
 H1 (+)
Profitability (ROE)

H2 (+)
Firm size (SIZE) Firm value
H3 (+) (Tobin’s Q)
Liquidity (CR)
H4 (-) H5 (+) H6 (+)

Capital structure (DER)

Figure 2.5: The proposed research model

Information:
: Partially : Through Moderator
 METHODS
3.1 DATA COLLECTION
Secondary data was collected from annual reports and financial reports
from the website https://cafef.vn/ from 2018 to 2022. The total number of
manufacturing firms on HOSE is 126 firms. However, after eliminating
manufacturing firms with insufficient data collection, the study will perform
analysis on the remaining data including 98 manufacturing firms and the
number of observations used in the study is 490 observations. In addition, data
on closing stock prices at the end of the year were collected from the website
https://www.hsx.vn/.
3.2 MEASUREMENT OF VARIABLES
Dependent variable
Firm value (Tobin’s Q): Firm value of manufacturing firms on HOSE.
This variable is measured by Tobin’s Q (Putro & Risman, 2021; Husna, 2020).
Tobin's Q is calculated by taking the market value of equity and adding it to
the total liabilities, which is then divided by the total assets (Klapper & Love,
2002). Specifically, this ratio can describe the performance of the stock market
over the book value. Tobin’s Q can be calculated using the following formula:
(EMV + D)
Tobin’s Q =
TA
Tobin’s Q is firm value;
EMV is the market value of equity (closing price x number of shares
outstanding);
D is the total debt;
TA is the total asset.
Sudiyatno & Pushpitasari (2010) explain Tobin's Q value as follows:
Tobin's Q < 1 indicates that the stock is undervalued. Low investment
growth potential.
Tobin's Q = 1 Describes the commodity in its average state. Investment
growth potential is not developed.
Tobin's Q > 1 illustrates that the stock is overvalued. The management is
successful in managing the company's assets. High investment growth
potential.
 Independent variable:
Profitability (ROE): Profitability is measured by the ratio between profit
after tax to total equity (Kasmir, 2010; Widnyana & Budiyasa, 2019;
Halfiyyah & Suriawinata, 2019). This ratio describes the level of a company's
ability to generate profits that can be realized by shareholders (Romansyah et
al., 2021). ROE is an indicator of how effectively management is using a
company’s assets to create profits. The formula for calculating ROE is as
follows: Profit after tax
ROE =
Total equity

Firm size (SIZE): Firm size is expressed through the total assets of the
company. Firm size can be calculated using the total asset logarithm (Ln)
(Romansyah et al., 2021; Husna, 2020). The formula for calculating SIZE is as
follows:
SIZE = Ln Total Asset
Liquidity (CR): Liquidity is a ratio to measure a company's ability to
meet its short-term obligations (Reschiwati et al., 2020). A company with high
liquidity means it can pay its short-term debts. Liquidity is the level of a
company's ability to meet its current obligations. Liquidity is measured by the
current ratio (CR) (Adiputraa & Hermawan, 2020; Nurwulandari et al., 2021).
CR is a ratio that describes the ratio of current assets to current liabilities. The
formula for calculating CR is as follows:
Current asset
CR =
Current Liability
Moderating variable:
Capital Structure (DER): Capital structure describes the ratio of total
debt to total equity, which is used to measure the extent to which a company
utilizes debt over its total equity. Moderating variable can strengthen or
weaken the relationship between the independent and dependent variables. In
this study, capital structure is a moderating variable. Capital structure is
measured by the debt-to-equity ratio (DER) (Romansyah et al., 2021;
Widnyana & Budiyasa, 2019). The formula for calculating DER is as follows:

Total Debt
DER =
Total Equity
 Table 3.1: Variables description
Label Explanation Measurement Expected sign Source
Dependent variable
Tobin’s Q Firm value The ratio of the sum of the market value of Umar et al., (2020); Al-Nsour &
equity and total debt divided total asset. Al-Muhtadi (2019); Sampurna &
Romawati (2019).
Independent variables
ROE Profitability The ratio between profit after tax to total + Sulastri et al., (2018); Romansyah
equity. et al., (2021); Ahmad & Muslim
(2022).

SIZE Firm size The logarithm (Ln) of the total asset. + Romansyah et al., (2021);
Sampurna & Romawati (2019).

CR Liquidity The ratio of current assets to current liabilities. + Mulyani et al., (2017); Ahmad &
Muslim (2022).
Moderating variable
DER Capital The ratio of total debt to total equity. +/- Sulastri et al., (2018); Romansyah
structure et al., (2021); Mulyani et al.,
(2017); Al-Nsour & Al-Muhtadi
(2019).
Source: Synthesizing from the author
3.3 ANALYSIS METHODOLOGY
3.3.1 Descriptive statistics
Descriptive statistics are used to describe the basic characteristics of the
data in a study. These are simple summaries of the sample and the methods of
measurement. Along with simple graphical analysis, they form the basis of
virtually any quantitative analysis of data. Descriptive statistics provide
general data including mean, variance, standard deviation, maximum, and
minimum.
The mean is the sum of the observed variables divided by the number of
observations. The mean can be calculated using the following formula:
n

X= ∑ Xi = X 1 + X 2 +...+ X 3
i
n
n
In which:
X is the mean;
X i is the value of observation i;

n is the number of observations.

Variance is the mean squared error between the variables and the mean
of those variables. Variance can be calculated using the following formula:
n

S2 =
∑ X i− X
i
n−1
In which:
S2: variance;
X : the mean value;
X i : the value of observation i;

n: number of observations.
Standard deviation: is the square root of the variance. Standard deviation
can be calculated using the following formula:
S = √ S2
In which:
S: Standard deviation;
 S2: Variance.
3.3.2 Correlation analysis
Pearson's correlation coefficient (r) is a statistical indicator used to
measure the correlation between two variables with a value from -1 to 1. The
test of the correlation coefficient is to test the linear relationship between the
independent and dependent variables. If the correlation coefficient is zero (or
close to zero), the two variables are not related; Conversely, if the absolute
value of the correlation coefficient is closer to 1, it means that the two
variables are highly correlated.
The correlation coefficient will be compared to the standards of Farrar
and Glauber (1967) of 0.8, if the correlation coefficient between variables does
not exceed 0.8, the model may not have a multicollinearity phenomenon.
3.3.3 Ordinary Least Squares, Fixed Effect Model, and Random
Effect Model
3.3.3.1 Ordinary Least Squares
The most widely used method for estimating parameters in regression
equations. To minimize the sum of squares of the vertical distances between
the collected data and the regression line (or surface) (Nguyen, 2006).
3.3.3.2 Fixed Effect Model
Assuming each observation has its characteristics that can affect the
explanatory variables, the fixed-effect regression method analyses the
correlation between the residuals of each observation and the explanatory
variables. Thereby, controlling and separating the effects of the individual
features (time constant) from the explanatory variables so that we can estimate
the effects of the explanatory variable on the dependent variable.
3.3.3.3 Random Effect Model
The random effect model (REM) is an alternative to the fixed effect
model (FEM), REM is a random constant term of regression analysis. It
depends on both a cross-section and a time series within them.
3.3.3.4 F-test
F test to select the OLS or FEM method, based on the last line of the
FEM model.
P – value < 0.05: Fixed effects model – FEM is selected;
 P – value > 0.05: Choose Ordinary least squares estimation model –
OLS.
3.3.3.5 Breusch and Pagan Lagrangian Test
Breusch and Pagan Lagrangian Test to test heteroskedasticity in the
model, based on the assumption:
P – value < 0.05: Random effects model – REM is selected;
P – value > 0.05: Choose Ordinary least squares estimation model –
OLS.
3.3.3.6 Hausman Test
The Hausman test is used to decide between a Fixed effect model and a
Random effect model. By running a Hausman test, you can determine which
model is best suited for your panel data analysis. If the null hypothesis is
accepted, it indicates that the random effect model is suitable, while accepting
the alternative hypothesis suggests that the Fixed effect model is more
appropriate for your analysis. The test examines whether the error terms are
correlated with the regression. Specifically, it tests the null hypothesis that the
Random effect model is suitable. If the test statistic is significantly large, the
Fixed effect model is preferred, whereas if the statistic is significantly small, a
Random effect model is preferred. This testing was conducted with the
following hypotheses:
H0: P – value > 0.05, then the Random effect model (REM) is valid to be
used;
H1: P – value < 0.05, then the Fixed effect model (FEM) is valid to be
used.
3.3.4 Autocorrelation, multicollinearity, and heteroskedasticity
testing
3.3.4.1 Autocorrelation
Autocorrelation is the phenomenon where at time t, the error has a
relationship with the error at time t - 1 or at any other time in the past.
3.3.4.2 Multicollinearity
Multicollinearity is the dependence of one independent variable on
another independent variable or multiple variables. This issue of
multicollinearity in the data hinders the precise measurement of the
relationship between the dependent and independent variables. To address
multicollinearity, the Variance Inflation Test (VIF) is commonly used. If the
value of VIF is less than 10, it indicates the absence of multicollinearity.
3.3.4.3 Heteroskedasticity
Heteroskedasticity refers to the difference in the variance of the error
term across observations. To determine if there is heteroskedasticity in the
data, one can examine the p–value and the constancy of the variable's
variance. If the p–value is significant and the variance of the variable is
constant, it indicates the absence of heteroskedasticity. The White test is
commonly used to detect heteroskedasticity. For the fixed effects model, the
Wald test is used, then the test value Prob>chi2 needs to be greater than 0.05
so that the research model does not have the phenomenon of
heteroskedasticity. For the random effects model, the Breusch and Pagan
Lagrangian test (Prob>chi2) needs to be greater than 0.05 so that
heteroscedasticity does not occur.
3.3.5 Generalized Least Squares
Generalized least squares estimation method when the model has
phenomena such as heteroskedasticity and/or autocorrelation. Then the GLS
method overcomes the above phenomena and these results will be used to
conclude and discuss the research results.
3.3.6 Moderating Regression Analysis
The moderating effect, also known as an interaction effect, occurs when
a moderator variable modifies the nature or strength of the relationship
between two or more variables (Hair et al., 2009). A moderating relationship
involves three types of variables: the dependent variable, the independent
variable, and the moderator variable. In this relationship, the independent
variable influences the dependent variable, and the presence of a moderating
relationship is indicated by variations in the influence of the independent
variable on the dependent variable based on the value of the moderator
variable.
In this research, the author uses specialized statistical software named
STATA14 for data analysis. From the testing results, the author will select the
appropriate regression model to implement the research. Therefore, the
proposed research model can be expressed as the following regression
equation:
Tobin’Qi,t = α + β1ROEi,t + β2SIZEi,t + β3CRi,t + β4(ROEi,t)*(DERi,t) +
β5(SIZEi,t)*(DERi,t) + β6(CRi,t)*(DERi,t) + ei,t
 In which:
Tobin’s Q is a dependent variable (firm value);
α is the intercept (the value of Tobin’s Q when the value of all variables
is usually 0);
β1 is the regression coefficient of profitability;
β2 is the regression coefficient of firm size;
β3 is the regression coefficient of liquidity;
β4 – β6 is the regression coefficients of the interaction of independent
variables with capital structure;
ROE is profitability;
SIZE is firm size;
CR is liquidity;
ROE*DER is the interaction of profitability with capital structure;
SIZE*DER is the interaction of firm size with capital structure;
CR*DER is the interaction of liquidity with capital structure;
DER is a moderating variable (Capital structure);
i are interviewed firms;
t is the survey period, 2018, 2019, 2020, 2021 and 2022;
ei.t is the error term of i firm at t time.
 RESULTS
4.4.2 Correlation analysis
Pearson's correlation coefficient (r) is a statistical indicator used to
measure the correlation between two variables, with a value ranging from -1 to
1. The test of the correlation coefficient is used to assess the linear relationship
between the independent and dependent variables. If the correlation coefficient
is zero (or close to zero), it indicates that the two variables are not related.
Conversely, if the absolute value of the correlation coefficient is closer to 1, it
indicates a high degree of correlation between the two variables. Additionally,
negative values represent a negative correlation.
The independent variable that has a positive correlation with firm value
is profitability, with a correlation coefficient of 0.406 and a significance level
of 1%. This shows that the firm has high profitability, providing a positive
signal for investors, thereby attracting investment and increasing firm value. In
addition, the independent variable that is correlated with firm value is firm
size, with a correlation coefficient of 0.329 and a significance level of 1%.
This shows that it is easier for large firms to attract investment capital than
small firms. The negative correlation between Tobin's Q and SIZEDER is -
0.151, with a significance level of 5%.
The results show that there is no serious multicollinearity phenomenon
(autocorrelation between independent variables in the model) because the
correlation coefficients have low values. In the model, the highest correlation
coefficient is 0.738. However, the value is still lower than the comparison
standard of 0.8 (Farrar & Glauber, 1967).
Table 4.4: Correlation matrix between variables

Tobin’s Q ROE SIZE CR ROEDER SIZEDER CRDER

Tobin's Q 1.000

ROE 0.406*** 1.000

SIZE 0.329*** 0.124*** 1.000

CR -0.028 0.013 -0.275*** 1.000

ROEDER 0.043 0.738*** 0.046 -0.027 1.000

SIZEDER -0.151*** -0.105** 0.251*** -0.258*** 0.197*** 1,000

CRDER -0.069 0.174*** 0.048 -0.025 0,537*** 0,598*** 1,000

***, **, * respectively indicates the significant level at 1% and 5%, and 10%
Source: Author’s data processing

57
 4.4.3 Analysing the impact of profitability, firm size, and liquidity on
firm value of manufacturing firms on HOSE with capital structure as a
moderating variable
After running the Ordinary least squares model, the author tested for
homoscedasticity and multicollinearity. The results of the homoscedasticity
and multicollinearity tests are shown in the following table:
Table 4.5: VIF and Homoscedasticity testing
VIF
White’s test
Variables VIF 1/VIF
ROE 2.85 0.351
SIZE 1.20 0.831
CR 1.16 0.864
chi2(24) = 118.50
ROEDER 3.61 0.277
SIZEDER 1.99 0.502
CRDER 2.37 0.422
Mean VIF = 2.20 Prob > chi2 = 0.000
Source: Author’s data processing
The VIF of all independent variables is less than 10, so the
multicollinearity phenomenon in the model is considered not serious.
White test result is: Prob > chi2 = 0.000, so Prob>chi2 < 0.05 should
reject the hypothesis H0, so the model has heteroskedasticity.

58
Table 4.6: Regression model of factor affecting the value of
manufacturing firms with capital structure as a moderating variable on
HOSE
OLS FEM REM
Coef. Coef. Coef.
Cons -3.923*** 4.378** -1.976*
(-6.10) (2.23) (-1.91)
ROE 4.453*** 1.509*** 2.234***
(13.21) (4.83) (7.33)
SIZE 0.167*** -0.118* 0.105***
(7.26) (-1.70) (2.86)
CR -0.001 -0.014 -0.010
(-0.19) (-1.30) (-1.16)
ROEDER -1.157*** -0.348*** -0.525***
(-9.42) (-3.62) (-5.46)
SIZEDER -0.004*** 0.003** 0.0002
(-2.73) (1.98) (0.19)
CRDER 0.086*** 0.009 0.028
(3.70) (0.51) (1.56)
R2 0.393 0.071 0.045
Number of observations 490 490 490
Autorcorrelation test p = 0.037
Heteroskedasticity test p = 0.000
F test F (97, 386) = 11.97, Prob > F = 0.000 < 0.05
Breusch and Pagan Chibar2 (01) = 316.82, Prob > chibar2 = 0.000
Lagrangian multiplier < 0.05
test
Hausman test Chi2 (6) = 125.29, Prob > chi2 = 0,000 < 0.05
***, **, * respectively indicates the significant level at 1% and 5% and 10%
Source: Author’s data processing

59
 Research model of profitability, firm size, and liquidity affecting the
value of manufacturing firms with capital structure as a moderating variable
on the Ho Chi Minh Stock Exchange in the period of 2018 to 2022. Based on
the results of table 4.6, Ordinary least squares (OLS), Fixed effect model
(FEM), and Random effect model (REM) were considered, and the model to
be used for the research was chosen as follows:
Firstly, the choice between Ordinary least squares model or Fixed effect
model was made using the F-test. The test results yielded a Prob > F value of
0.000, less than 0.05, with F (97, 386) = 6.84. This result shows that the fixed
effect model is more suitable than the Ordinary least squares model.
Secondly, the choice between Ordinary least squares model or Random
effect model was made using the Breusch and Pagan Lagrangian multiplier
test. The test result showed a Prob > chibar2 value of 0.000, less than 0.05,
with chibar2 (01) = 151.44. This value indicates that the suitable model for the
research is the random effect model, and the random effect model was
selected.
Finally, the choice between the fixed effect model or Random effect
model was made using the Hausman test. The test result yielded Prob value >
chi2 of 0.000, which is less than 0.05, with Chi2 (6) = 108.47. This provides
enough evidence to confirm that the fixed effect model is more suitable than
the random effect model. Thus, the fixed effect model will be used for
subsequent analysis.
The estimates from the Fixed effect model (FEM) model are used to
explain the effects of profitability, firm size, and liquidity on firm value, with
capital structure as a moderating variable. The estimated results using Fixed
effect model (FEM) show that profitability has a positive impact on firm
value, while firm size has a negative impact on firm value. In addition, capital
structure weakens the positive impact of profitability on firm value, while
capital structure weakens the negative impact of firm size on firm value.
Autocorrelation and heteroscedasticity tests were performed to detect
model defects. The determined results are presented as follows:
Autocorrelation test results
To check for the autocorrelation phenomenon in panel data, the
Wooldridge test was performed with the null hypothesis (H0) stating that there
is no autocorrelation. By executing the 'xtserial' command in STATA, the

60
obtained result was p = 0.037 < 0.05. Therefore, the author rejects hypothesis
H0 and concludes that there is an autocorrelation phenomenon.

61
 Heteroscedasticity test results
The modified Wald test was conducted to test for the heteroscedasticity
phenomenon, with the null hypothesis (H0) stating that there is no
heteroscedasticity. By executing the 'xttest3' command for the FEM model, the
resulting p-value of p = 0.000 < 0.05 Therefore, the author rejects hypothesis
H0 and concludes that there is a heteroscedasticity phenomenon in the model.
Based on the results of testing autocorrelation and heteroscedasticity, it
can be concluded that the model has an autocorrelation and heteroscedasticity
phenomenon. Therefore, the author employs GLS model to overcome the
regression model's defects, such as autocorrelation and heteroscedasticity.
Table 4.7: Generalized least squares model results
Variables Coef. Std. Err P-value
Cons -1.150 0.452 0.011
ROE 2.695 0.198 0.000
SIZE 0.067 0.016 0.000
CR -0.003 0.002 0.038
ROEDER -0.638 0.065 0.000
SIZEDER 0.001 0.001 0.153
CRDER 0.032 0.012 0.007
Number of observations 490
P-value 0.000
Source: Author’s data processing
Table 4.7 is the result of the GLS model, which is an appropriate model
to explain the influence of profitability, firm size, and liquidity on firm value
of manufacturing firms with a capital structure as a moderating variable on
HOSE. With the GLS model, a p-value = 0.000 < 0.05 represents the
significance of the model; this value proves that the model is suitable for
considering the impact of variables.
In summary, based on the results of table 4.7, it can be seen that
profitability and firm size have a positive impact on firm value with
coefficients β = 2.695 and β = 0.067, respectively, with the same significance
level of 1%, while liquidity has a negative effect on firm value at the 5%
significance level (β = -0.003, p < 0.038). Further, capital structure weakens

62
the positive impact of profitability on firm value at the 1% significance level
(β = -0.638, p < 0.000), but capital structure weakens the negative impact of
liquidity on firm value at the 1% significance level (β = 0.032, p < 0.007).
Moreover, capital structure does not moderate the impact of firm size on firm
value.
4.4.4 Result and discussion
Independent variables
Profitability (ROE): Profitability is measured by the ratio of profit after
tax to total equity. This variable has a statistical significance level of 1%, with
a regression coefficient β = 2.695. The positive regression coefficient indicates
that profitability has a positive effect on firm value. This result is following
the study's expectations; hypothesis H1 is accepted. When profitability
increases by 1 unit, firm value increases by 2,695 units. Profitability has a very
strong influence on firm value. The results are consistent with signaling
theory, profitability brings a positive signal to investors when the company is
highly profitable. High profitability reflects a large company's profits, which
indicates that the company's operating efficiency is good. On the other hand,
the higher the profits the company earns, the more likely it is to pay dividends
to investors, which leads to increased investor confidence and increased stock
prices and firm value. Investors often rely on this ratio to compare the stocks
of different firms and make decisions about which company's stock to buy.
High profits will be a sign of the company's good prospects to attract investors
to participate in the growing demand for buying shares (Rahayu & Darim,
2020). This result is consistent with some previous studies on profitability
having a positive effect on firm value (Hermuningsih, 2014; Sari &
Witjaksono, 2021; Agustina, 2013). Besides, the results of this study do not
support Panjaitan & Supriyati (2023) and Kristianti & Foeh (2020), who state
that profitability had a negative impact on firm value. In the Vietnamese
context, some previous studies have shown that profitability had a positive
impact on firm value (Pham & Pham, 2017; Le et al., 2021).
Firm size (SIZE): Firm size is expressed through the total assets of the
firm. Firm size has a statistical significance level of 1%, with a regression
coefficient β = 0.067. The positive regression coefficient indicates that firm
size has a positive effect on firm value. This result is following the study's
expectations; hypothesis H2 is accepted. When firm size increases by 1 unit,
firm value increases by 0.067 units. Large firms will more easily attract
investment and cooperation from domestic and foreign firms than small firms,

63
leading to an increase in company assets to serve firm activities. At the same
time, large firms make investors trust and feel that the company will bring
more benefits than small firms, so the stock price increases, increasing the
value of firms. Large firms have greater flexibility in obtaining the necessary
capital to realize profitable investment opportunities than small firms
(Romansyah et al., 2021). In the same context as the Vietnamese stock market,
research by Vo (2017); Nguyen & Truong (2016); and Nguyen (2023) showed
that firm size has a positive impact on firm value. In addition, the results are
also consistent with the studies of Lambey et al., (2021); Pratiwi et al., (2022);
Yusuf et al., (2023). The results of this study do not support Bon & Hartoko
(2022) and Setiadharma & Machali (2017), stating that firm size does not
affect firm value.
Liquidity (CR): Liquidity represents a firm's ability to fulfill short-term
obligations. The finding shows that there is a negative relationship between
liquidity and firm value in the GLS model, with a regression coefficient β =
-0.003 at the 5% statistical significance level. The negative regression
coefficient indicates that liquidity has a negative effect on firm value, which
rejects hypothesis H3; when liquidity increases by 1 unit, firm value decreases
by 0.003 units. This shows that the higher liquidity, the lower firm value. The
results show that the existence of a large amount of short-term assets (cash,
accounts receivable, inventory), which are not used by managers in business
activities, leads to increased liquidity and a decrease in firm value. This
observation is consistent with collected data, the majority of manufacturing
enterprises have a large amount of short-term assets compared to short-term
liabilities, and the firm has this ratio reaching 52.26 times. According to
research by Do & Pham (2021) and Nguyen et al., (2020) in the context of
Vietnam’s stock market, the results of the study show that liquidity has a
negative impact on firm value but is not statistically significant. Besides,
foreign studies show that firms with higher liquidity tend to have lower firm
value (Hidayah & Rahmawati, 2019; Rizky & Kisman, 2023). The results of
this study do not support Hapsoro & Falih (2020) and Fadjri & Nurdiansyah
(2023), who state that liquidity has a positive impact on firm value.
Moderating variable
Based on the results in table 4.6, the results in the GLS model show that
capital structure weakens the positive impact of profitability on firm value,
with the regression coefficient β = -0.638 at the 1% statistical significance
level. The negative regression coefficient shows that if the capital structure is
higher, the positive effect of profitability on firm value will decrease. When

64
the interaction value of profitability and capital structure increases by 1 unit,
the firm value will decrease by 0.638 units. The positive impact of profitability
on firm value will be reduced because the level of risk associated with the
capital structure makes investors afraid. A high capital structure shows that the
company uses a lot of debt and has to pay interest costs from debt, leading to a
decrease in the benefits investors receive from the company, so it will be
difficult for investors to make decisions when choosing to invest in such a
firm. Wansani (2023) also argues that the benefits of borrowing will become a
burden for the firm when it exceeds the optimal point, so the company has to
bear debt and interest, and the capital structure weakens the relationship
between profitability and firm value. Bonansius (2022); Fauziah & Sudiyatno
(2020); and Fatima et al., (2023) show that capital structure weakens the
positive impact of profitability on firm value. In addition, the results of this
study do not support Izzah et al., (2023) and Diastanova and Marsoem (2023),
who state that capital structure cannot moderate the impact of profitability on
firm value.
According to the results in the GLS model, the regression coefficient β =
0.001 with a p-value = 0.153 greater than 10% indicates that this variable is
not statistically significant. Thus, there is no evidence that capital structure
moderates the relationship between firm size and firm value. This result shows
that when the capital structure increases, it does not moderate the positive
impact of firm size on firm value. The stability of large firms is attractive
because the management of large firms will operate optimally to maintain the
company's status, the stable status of the company will attract investors, which
can automatically increase the stock price. Tarigan et al., (2022) and Afni et
al., (2023) did not find a moderating role of capital structure on the impact of
firm size on firm value. The results of this study do not support Izzah et al.,
(2023) and Mardevi et al., (2020), who state that capital structure strengthens
the impact of firm size on firm value, while firm size does not directly impact
firm value.
Based on the results in table 4.6, the results in the GLS model show that
capital structure weakens the negative impact of liquidity on firm value, with
the regression coefficient β = 0.032 at the 1% statistical significance level. The
positive regression coefficient can be understood as the negative impact of
liquidity on firm value decreasing when moderated by a higher capital
structure. As the value of capital structure increases in the interaction between
liquidity and capital structure by 1 unit, the firm value will increase by 0.032
units. A high capital structure will help firms solve problems with capital

65
sources, receivables, and inventory, which cause the negative effect of
liquidity on firm value. Besides, a reasonable capital structure will give
investors confidence in their investments and the company's ability to use cash
to create benefits for investors, thus increasing the value of the company. In
Vietnam, manufacturing firms often have high inventories and high
receivables from customers, so a suitable capital structure will help regulate
these issues for firms, increasing firm value. The ability to pay debt represents
the stable capital structure of the company, indicating that the company has
good operating performance (Zulfa et al., 2022). This research result is
consistent with the study of Indira & Wany (2021). The results of this study do
not support Hasanudin et al., (2022), Afni et al., (2023), and Mardevi et al.,
(2020), stating that capital structure cannot moderate the effect of liquidity on
firm value. Furthermore, Rizky & Kisman (2023) argue that capital structure
weakens the positive effect of liquidity on firm value.

66
 APPENDICES
1. Statistics describe the variables in the model

2. Correlation matrix between variables in the research model

67
3. Results of analyzing the influence of profitability, firm size, and
liquidity on firm value with capital structure as moderating variable:
Evidence of manufacturing firms on HOSE
3.1 Ordinary Least Square (OLS)

3.2 Multicollinearity test

68
3.3 Test for heteroskedasticity

3.4 Fixed effect model (FEM)

69
 3.5 Random effect model (REM)

3.6 Breusch test and Pagan Lagrangian multiplier test to choose OLS or REM

70
3.7 Hausman test to choose FEM or REM

3.8 Test for heteroskedasticity

3.9 Test for autocorrelation

71
3.10 GLS model estimation results

72