Nghiên cứu tác động của công bố thông tin trách nhiệm xã hội đến hiệu quả tài chính của các doanh nghiệp niêm yết trên thị trường chứng khoán việt nam tiếng anh

1

INTRODUCTION
1. Background
Corporate social responsibility (CSR) is a part of non-financial information showing
interaction of enterprises with society and living environment. Through this information,
stakeholders can assess the level of practice of social responsibility of enterprises. So
corporate social responsibility disclosures (CSRD) play a significance role in promoting
an enterprise’s image with related parties to enhance competitive capacity, attract
investment, improve the financial performance of enterprises. Many companies in the
world have had certain achievements through practice and disclosures of CSR. For
instance, Microsoft has become the leading prestigious technology company in the world
surveyed by Reputation Institute in 2012 with community support programs, charity
programs. Another example is the revival of Nike after a sales slump due to accusations
and lawsuits related to scandals involving Nike’s suppliers exploiting workers with cheap
wages. This issue revealed by the publicity of Nike’s suppliers was monitored by nongovernmental organizations in 2004. Indeed, achievements from practice and disclosures
of CSR have also been proved by scientific works of researchers in the world that
practice and disclosures of CSR can help to enhance credibility of enterprises (Hess et al.,
2002; Brammer and Millington, 2005; Yingjun Lu et al., 2015), increase market value
(Belkaoui, 1976; Frankle and Anderso,1980; Robert,1978; Martin Freedman and
Stagliano, 1991; Berthelot et al., 2012; Clarkson et al., 2013; Klerk et al., 2015, Cahan et
al., 2015), reduce capital costs (Dhaliwal et al., 2011; Orens et al., 2010), enhance
employee satisfaction (Kim et al., 2010). Thanks to these benefits, the practice and
disclosures of CSR has become a global trend supporting survival and sustainable
development of enterprise.
Vietnam is not an exception to this global trend as the Government and enterprises
have placed an increasing attention to CSR with the purpose of sustainable development.
The Government also provides mechanisms to support and guide enterprises to
implement and report social responsibility activities to aim at sustainable development.
For example, in 2012 the State Securities Commission in conjunction with a number of
organizations such as VCCI, IFC, ACCA, GRI issued guidelines for developing

sustainability reports based on global reporting standards. 2015 is the marking year of the
increasing pressure of the State towards Vietnamese enterprises when the establishment
of the Circular No. 155/2015/TT-BTC dated October 6, 2015 of the Ministry of Finance
was considered as the first legal document requiring disclosure of sustainable
development of listed companies. Regarding Vietnamese enterprises, in recent years,
social responsibility activities have been partly integrated into production and business

2

activities but mostly in listed companies on the stock market. Some listed companies
have been pioneer in CSRD even when there are no legal regulations required to prepare
sustainable development reports such as Bao Viet Group, Vietnam Dairy Products Joint
Stock Company and Ho Chi Minh City Securities Corporation, Hoa Sen Group Joint
Stock Company.
However, the question is whether all Vietnamese enterprises are interested in and
implement CSR and whether the practice and disclosures of CSR actually bring benefits
to Vietnamese enterprises as evidence shown in the world. In order to figure out the
answer, many researchers have also conducted surveys on Vietnamese enterprises and
provided evidence that CSR enhances financial performance (Chau Thi Le Duyen, 2013;
Tran Thi Hoang Yen, 2016), increase customer loyalty (Nguyen Thi Anh Binh and Pham
Long, 2015), affect buying behaviors of consumers (Nguyen Phuong Mai, 2015).
Furthermore, there has been some research on CSRD showing that CSRD affects
corporate financial performance such as studies of Ho Ngoc Thao Trang and Liafisu Sina
Yekini (2014), Nguyen Thi Bich Ngoc et al. (2015), Nguyen Xuan Hung and Trinh Hiep
Thien (2016), Ta Thi Thuy Hang (2017), Ho Thi Van Anh (2018). However, these studies
have inconsistent results and small research samples with limited observations, short
research time and only use common multiple regression (OLS, FEM, REM) to understand
the impact of CSRD on financial performance of enterprises but could not resolve the
endogenous phenomena in the research model as CSRD and financial performance
variables may have reciprocal relationship. At the same time, practice of CSR in Vietnam

is increasing, but no research has been conducted to determine whether the increasing
practice of CSR can lead to the increase of financial performance of enterprises. Therefore,
it is critical to use the additional model estimation method to overcome the endogenous
phenomenon between variables and looking for differences over time in the impact of
CSRD on financial performance to confirm reliable and complete conclusions concerning
impact of CSRD on financial performance of enterprise
For all the above reasons, it is necessary to expand previous studies on the impact of
CSRD on the financial performance of enterprises, to come into clearer and more reliable
conclusions. Therefore, this paper aims at investigating “The impact of corporate social
information disclosure on financial performance of listed companies on Vietnam's
stock market”
2. Objectives and research questions
2.1. Objectives
As stated above, the main purpose of this study is to investigate the impact of
CSRD on financial performance of listed companies on Vietnam's stock market.


3

2.2. Research questions

4

Chapter 5: Discussion, recommendations and conclusion
CHAPTER 1: LITERATURE REVIEW

Based on the research, this study aims at answering following questions:
1. Is there any difference between CSRD of manufacturing enterprises and nonmanufacturing enterprises on the Vietnam's stock market?
2. Do CSRD affect the financial performance of listed companies on the Vietnam’s
stock market? Is there any difference in different time periods? Does CSRD of certain

year affect the financial performance in the following year?

1.1. Review of previous studies on the impact of corporate social responsibility
disclosures on corporate financial performance
1.1.1. CSR was approached in different ways by researchers with varied contents
1.1.2. Corporate financial performance was measured by different indicators by
researchers

3. Research subjects and scope

Research measuring corporate financial performance by market value

3.1. Research subjects:

Research measuring corporate financial performance by accounting data

Level of CSRD, financial performance of enterprises, impact of CSRD on financial
performance of enterprises listed on the Vietnam’s stock market.
3.2. Research scope
+ Content: In the research, CSRD is measured based on the level CSRD and
financial performance of enterprises is measured by two indicators which are return on

Overview of studies on the impact of CSRD on corporate financial performance showed
that most research found a positive relationship between these two variables. However, there
are some studies showing negative or no relationship between these two variables. This is
explained by some of the following reasons:
First, differences in research contexts

assets ROA and the market value Tobin’Q. The content of CSR is a combination of 4
information components: Information of environmental responsibility, Information of

labor responsibility, Information of community responsibility, Information of consumer
responsibility.

Second, differences in methodology

+ Subjects: Listed companies disclosing non-financial information on the stock
market in both Ho Chi Minh and Hanoi. Enterprises listed on the Unlisted Public
Company Market (UPCOM) are not covered in the thesis.

Fifth, differences in research time

+ Time: Data of the companies listed on the Ho Chi Minh City and Hanoi Stock
Exchange are publicized from 2006 to 2016. The companies listed after 2006 and delisted
during the period from 2006 to 2016 are not covered in this thesis.
4. Structure
Except introduction, conclusion, appendices and references, the thesis has 5 main
chapters:
Chapter 1: Literature review
Chapter 2: Theoretical foundation of the impact of corporate social responsibility
disclosures on corporate financial performance.
Chapter 3: Methodology
Chapter 4: Research results

Third, differences in control variables of research models
Fourth, differences in measurements of CSRD and corporate financial performance

1.2. Research gap
Thus, based on literature review of studies in Vietnam and abroad, there exist following
research gaps:
Firstly, research context and time can be factors influencing results of studies on the

impact of CSRD on corporate financial performance.
Secondly, the majority of studies measured the level of CSRD based on the method
of content analysis. However, there exist differences among studies and reliability of scales
remains controversial.
Thirdly, diversity in using indicators of corporate financial performance can lead to
inconsistent conclusions about the impact of CSRD on corporate financial performance.
Fourthly, previous studies use common multivariate regression (OLS, FEM, REM)
to investigate the impact of CSRD on corporate financial performance but fail to take into
account endogenous phenomenon in the research model as CSRD and corporate financial
performance may have reciprocal relationship.


5

CHAPTER 2: THEORETICAL BACKGROUND OF THE IMPACT OF
CORPORATE SOCIAL RESPONSIBILITY DISCLOSURES ON CORPORATE
FINANCIAL PERFORMANCE
2.1. Theories of corporate social responsibility and corporate social responsibility
disclosures CSR
2.1.1. Corporate social responsibility

6

Firstly, United Nations Global Compact (UNGC)
UNGC provide a fundamental direction for enterprises to establish, implement and
disclose information of sustainable policies and responsible actions. UNGC wants
companies to integrate in their production and business strategies according to 10
principles set by UNGC. Furthermore, enterprises and member organizations are required
to prepare annual Communication of Progress to display actual activities that enterprises
implemented or had plans to implement 10 principles of UNGC.

Social responsibility is a complex concept which is defined according to many different
standards. Although definitions of CSR may be different, the main concept of CSR is that in
addition to the development of each enterprise in accordance with the current law,
companies’ activities should be associated with the common interests of the social
community.
2.1.2. Corporate social responsibility disclosures

Secondly, Global Reporting Initiative (GRI)
GRI provides guidelines for voluntary application of sustainable development reports
for all types of companies. This standard sets out reporting principles as well as economic,
environmental and social indicators. In the latest version G4 released in 2013 GRI provides
a guide to disclose information according to 3 groups: economic, environmental and social
category.

2.1.2.1. Nature of corporate social responsibility disclosures
Thirdly, International Standards Organization (ISO) 26000:2010
CSRD derives from CSR. Specifically, CSRD is a critical tool used by enterprises
to communicate information of implementation of corporate social responsibility to
stakeholders. Rob Gray et al. (1995) supposed that CSRD is a process providing
information of social and environmental impact caused by economical activities to
interested parties.

ISO 26000 provides guidelines for voluntary implementation of corporate social
responsibility for all organizations in both public and private sectors, in developed and
developing countries, as well as transition economies. This set of standards identifies the
core objectives of CSR with 7 categories: corporate governance, human rights, labor
practices, the environment, fair practices, consumer issues, community development.

2.1.2.2. Methods of disclosing corporate social responsibility

Fourthly, International integrated reporting Council (IIRC)
There are many channels and ways for enterprise to implement disclosing
information such as meetings, public events, forums, reports, news, magazines, posters,
advertisements, electronic mails, videos, websites, podcast, blog, packaging, interviews.

Integrated reports are established on the basis of available financial statements,
combined with non-financial criteria. Enterprises can choose to publish sustainability
reports independently and/or integrate them in the content of annual reports.

2.1.3. Measurement of corporate social responsibility disclosures
2.2. Financial performance and measurement of financial performance
Level of CSRD is measured by amount of information through counting symbols,
number of words, sentences, pages, and ratio of social responsibility over total information a
company public on reports.
CSRD is measured by indicators. Accordingly, researchers develop a list of different
aspects of social responsibility to get a broad picture of practice of CSR. Based on the list,
researchers conduct indicator of CSRD by labeling CSR with numbers.

2.2.1. Financial performance
Performance: According to British – English dictionary performance is defined as
activities or processes of performing or completing specific tasks and work. Performance
concept is used in all areas of life, economy and society (from production and business to
health, education, defense...). It reflects the level of human using necessary elements to
participate in activities to achieve desired results.

In addition to the above content analysis method, some studies use assessment of
CSRD of a third party to assess the level of CSRD.
2.1.4. Standards for corporate social responsibility disclosure and regulations in
Vietnam

In the course of business, Venkatraman and Ramanujam (1987) supposed that
corporate performance reflects context of managing strategies. Regarding small scale,
performance of enterprises is measured by financial indicators reflecting the completion
of economic objectives of enterprises. These indicators reflect financial performance


7

(Venkatraman and Ramanujam,1987). These indicators may be indicators reflecting
revenue growth, profitability (ROA, ROE, ROI), earnings per share (EPS) or indicators
measured through market prices such as stock prices, market to book values or its
variants.
2.2.2 Measurement of financial performance

8

2.3.3. Signaling theory
Signaling theory was introduced by Michael Spence in 1974 in the context of
imperfect market and asymmetry of information. Information asymmetry arises
between those parties who hold information and those who are able to make better
decisions if they have this information.
CHAPTER 3: METHODOLOGY

The first group is traditional financial indicators
Traditional measures are used to evaluate financial performance based on accounting
databases. According to Nguyen Thi Thanh Hai (2015), there are four groups of basic
groups of indicators to assess financial performance of enterprises: the first group reflects
profitability, the second group reflects the efficiency of using assets, the third group
reflects solvency and the fourth group reflects operating capital.

3.1. Hypothesis
Based on the stakeholder theory, legitimacy theory and signaling theory, following
hypothesis is proposed:
H1: Corporate social responsibility disclosures has a positive impact on corporate
financial performance

The second group is modern financial indicators
3.2. Conceptual framework
Financial performance of enterprises can be measured based on market values such
as Earnings Per Share (EPS), Economic Value Added (EVA), Price to Earnings (P/E),
Market Value Added (MVA), market value Tobin'Q. These indicators are more interested
by investors because these indicators are directly related to investors’ interests to make
decisions to buy stocks or invest in other sectors.
2.3. Theoretical background of the relationship between corporate social
responsisbility disclosure and corporate financial performance
2.3.1. Stakeholder theory

In this study, quantitative method is used to examine the impact of CSRD on
control variables size, financial leverage and growth rate of enterprises.
CSRD

FP

EMP

TBQ

COM
CUS
ENV

can be generated.
2.3.2. Legitimacy theory
Guthrie and Parker (1989) provided that enterprises operating in society must sign a
social contract in which managers agree to fulfill certain social requirements to achieve
their goals. The terms of this contract may be obvious such as requirements of the law but
may not be clearly defined depending on the expectations of the social community with
enterprises.

SIZE
LEV

Freeman (1983) provided that if an enterprise only care about interests of
shareholders and ignore needs of other stakeholders - those who can affect or be
affected in the course of achieving corporate goals, that enterprise may have to close
down. Freeman concluded that the goal of businesses is to meet the needs of
stakeholders who are affected by decisions of enterprises, if this is achieved, profits

ROA

VARIABLES

GRW

Figure 3.1: Conceptual framework
Corporate social responsibility disclosures (CSRD)
Corporate social responsibility is classified into 4 most typical and popular groups
which receive the most concerns from stakeholders in Vietnam including: Information on
environmental responsibility (ENV), information on responsibility to employees (EMP),
Information on responsibility to community (COM), information on responsibility to

customers (CUS).
Corporate financial performance
In this study, two indicators were used to measure corporate financial performance,
namely: Return on Assets (ROA), market value (Tobin'Q) to show the diversity of
indicators.


9

Control variables
In this study, control variables include corporate size (SIZE), financial leverage
(LEV) and revenue growth (GRW) as these variables were used by many researchers
when analyzing the impact of practice and disclosures of corporate social responsibility
of enterprises.
3.3. Quantitative research methods
To test the hypothesis mentioned above with the variables identified in the
research model, the author builds the regression equation as follows:
The following hypothesis is tested: CSRD has a positive impact on corporate
financial performance which is reflected in the following regression equation:
Model 1: ROAi,t= βo + β1CSRDi,t + β2SIZEi,t + β3LEVi,t+ + β4GRWi,t+ ui,t
Model 2: TBQi,t= βo + β1CSRDi,t + β2SIZEi,t + β3LEVi,t + β4GRWi,t+ ui,t
The impact of corporate social responsibility disclosures of previous year on
financial performance of the following year is calculated by two regression equations as
follows:
Model 3: ROAi,t= βo + β1CSRDi,t -1+ β2SIZEi,t + β3LEVi,t+ β4GRWi,t+ ui,t
Model 4: TBQi,t = βo + β1CSRDi,t -1+ β2SIZEi,t + β3LEVi,t+ β4GRWi,t+ ui,t
The impact of corporate social information disclosure on financial performance
taking into account changes over time is estimated as follows:
Model 5: ROAi,t= βo+β1CSRDi,t+β2CSRDi,t*Di,t + β3SIZEi,t + β4LEVi,t+β5GRWi,t+

10

Specific steps of the process of analyzing content of enterprises’ reports are
conducted through the following steps:
Step 1: Establish a temporary list of indicators of corporate social responsibility
Four groups of CSR: environment, employees, community connection, products
are established based on several sources. Specifically, those groups are inherited from
a study of Gunawan (2007) (Gunawan et al., 2008)) and Jitaree (2015). In addition,
they come from 3 documents in Vietnam: firstly, the guideline to prepare
sustainability reports 2012 (BCBV 2012) issued by the State Securities Commission
and the International Financial Organization; secondly, Circular 155/2015/TT- BTC
(TT 155) dated 06/10/2015 on guidelines for disclosures of information on the stock
market; thirdly, a study of Nguyen Thi Kim Chi (2016). Based on these documents, 35
indicators of information was established and classified as four main groups:
Information on environmental responsibility – ENV (10 indicators), information on
responsibility to employees – EMP (11 indicators), information on responsibility to
the community – COM (6 indicators) and information on responsibility to customers –
CUS (3 indicators).
Step 2: Conduct a pilot test to establish a complete list of indicators of CSR
A pilot test was conducted to check the appropriateness of the indicators of
information established in step 1. After the test, adjustment and supplement were made so
that the list of indicators are more suitable to disclosures of information of listed
companies on Vietnam stock market.

ui,t

Based on the results, a complete list of indicators of CSR was established to identify
the level of CSRD. The author with the support of assistants assessed level of CSRD by
analyzing content of annual reports and sustainability reports of enterprises which
disclose CSR

In which D is a dummy variable with value of 0 in the years 2006 and 2007, and
with value of 1 since the year 2008

After determining the level of corporate social responsibility disclosures (CSRD),
total level of CSRD of each enterprise in each year is determined as following formula:

ui,t
Model 6: TBQi,t= βo+β1CSRDi,t+β2CSRDi,t*Di,t + β3SIZEi,t + β4LEVi,t+β5GRWi,t+

3.4. Measurement of variables in the research model
3.4.1. Corporate social responsibility disclosure
In the context of Vietnam, the results of indicators of CSR are not available.
Therefore, in this study, the content analysis method is used in reports disclosing CSR
(annual reports, sustainability reports), thereby identify the level of CSRD by converting
qualitative information into quantitative information.

CSRDj=

ij

In which:
CSRDj: Indicator of information disclosures of the jth enterprise
Xij = 0 if ith indicator of information is not disclosed in the jth enterprise
Xij = 1 if ith indicator of information disclosed in the jth enterprise is general
information or quantitative information without specific explanation

(3.1)

11

12

Xij = 2 if ith indicator of information disclosed in the jth enterprise is detailed
information about specific activities

The final research sample consisted of 43 companies listed on the stock market of
Vietnam excluding financial companies, credit institutions banks, companies delisted
during the research period, and companies that the author could not find annual reports or
sustainability reports. Thus, with 43 enterprises during the period of 11 years, the author

3.4.2. Measurement of financial performance
Return on assets ROA. This indicator measures how well a company is generating
profits from its total assets. This indicator is identified by the following formula:

could collect all necessary data with total number of observations of 43×11=473.
Enterprises in the sample are divided into two groups: 27 manufacturing enterprises and
16 non-manufacturing enterprises.
Data

Profits before taxes
ROA

=

(3.2)
Average total assets

Profits before taxes are used as corporate income taxes of Vietnam during the

period from 2006 to 2016 varied: 28%, 25%, 22%, 20%.
Corporate market value Tobin’Q (TBQ): TBQ is a modern financial measuring
tool introduced by James Tobin in 1971 to the market value of total assets compared to
book value of total assets (Tobin, 1971).
In this study, TBQ is calculated as follows:
Market value VCSH + Book value of liabilities
TBQ =

(3.3)
Book value of total assets

In which: market value VCSH = Share price x the number of outstanding shares

Financial data for calculating indicators of the research was collected directly from
the financial statements of enterprises.
Data of CSRD was collected directly by analyzing the content of annual reports and
sustainability reports in Vietnamese version of enterprises in the sample which was
collected from the page />3.6. Research method
In this study, the author used panel data in regression analysis. Panel data has many
advantages over time-series data and spatial data (cross-section data) because it allows
studying of complex models fluctuating in both directions. In this study, author used the
Ordinary Least Squares regression (Pooled OLS), Fixed effects model (FEM), Random
effects model (REM), The Generalized Least Square (GLS), Two-stage least squares
(2SLS), Generalized method of moment (GMM) regression to find the best model for
research data.

3.4.3. Measurement of control variables
Abbreviation

Description

CHAPTER 4: RESULTS

Formula

SIZE

Enterprise size

Ln (assets)

LEV

Leverage

Total liabilities /total capital

GRW

Revenue growth

4.1. Characteristics of the Vietnam stock market and listed companies on the
Vietnam stock market
4.2. Descriptive statistics

Revenue of year (t ) – revenue of year (t-1)

4.2.1. Descriptive statistics of corporate social responsibility disclosures

Revenue of year (t-1)

4.2.1.1. Descriptive statistics of corporate social responsibility disclosures of the sample
3.5. Research data
Sample

The statistic results of the level of CSRD of 43 listed enterprises during the period
from the 2006 to 2016 showed that the average level of CSRD increased steadily. This
indicates that enterprises have placed increasing concern on implementing and disclosing
CSR.


13

14

Detailed information on average levels of CSRD on environment (ENV),
responsibility to employees (EMP), responsibility to community (COM), responsibility to
customers (CUS) indicates that the level of environmental disclosure is the lowest.

4.2.2. Descriptive statistics of ROA and TBQ

Specifically, by examining each indicator of CSR of total of 473 observations (11

ROA: Return on assets. During the period from 2006 - 2016 there is a significant
difference between enterprises. In 2009 ROA of enterprises has the highest value of
14.5%. In 2008 and 2014, ROA has the lowest value of 9.8% and 9.6% respectively.

year x 43 enterprises), it is found that information about salaries, bonus, production
process and developing products is published the most by enterprises. Information
disclosed the least is preventing and resolving environmental consequences caused by

TBQ: Corporate market value. This variable also varies between enterprises during
the period from 2006 to 2016. This variable reflects the attractiveness of corporate shares.
In 2011, the average value of TBQ has the lowest value of 0.955 because 2011 is

enterprises.

considered as the year of the strongest decline of the Vietnam stock market. In 2006 and
2007, TBQ has the highest value of 2.64 and 2,502 respectively. This is explained by the
fact that 2006 and 2007 is the booming year of the Vietnam stock market.

In another aspect, level of concern of 43 enterprises in the sample on CSRD was
examined through reporting activities. The results showed that from 2006 - 2016, the
number of enterprises implementing CSRD as a separate item in the report is very small.

4.3. Correlation analysis
Correlation analysis showed that:
- ROA has a significant positive correlation with independent variable CSRD and
negatively correlated with control variable LEV. This shows that the higher the level of
CSRD, the higher the efficiency of using assets and the higher the rate of liabilities over
capital, the lower the efficiency of using assets.

Figure 4.4: listed enterprises disclosing CSRD separately from 2006 to 2016
Source: calculations of the author
4.2.1.2. Descriptive statistics of corporate social responsibility in manufacturing

- TBQ has a significant positive correlation with independent variable CSRD and
insignificant correlation with other variables. this indicates that the higher the level of
CSRD, the higher the corporate market value.
4.4. Regression analysis

and non-manufacturing enterprises

The regression analysis is conducted in as following order:

By examining the level of CSRD in manufacturing and non-manufacturing
enterprises, it is proved that the levels of CSRD in manufacturing enterprises and non-

Step 1: Selecting Fixed effect and Random effect

manufacturing enterprises are significantly different.

Step 2: Comparing model selected in step 1 and model Pooled OLS to select the
best model
Step 3: Checking limitations of model selected in step 2
The author conducted tests to check limitations of the model: multi-collinearity,
heteroskedasticity, auto-correlation and endogenous phenomenon
If the multi-collinearity exists, one of two variables having multi-collinearity from
the model. If endogenous phenomenon exists, suitable adjustment should be made to

Figure 4.5: Levels of CSRD in manufacturing and non-manufacturing enterprises
during the period from 2006 to 2016
Source: calculations of the author

reduce limitations of the model.
Limitations of the models are displayed in the following table:
Table 4.12: Limitations of research models


15

Model

Dependent

Auto-

variable

correlation

16

Heteroskedasticity

Endogenous
phenomenon

model 4 showed positive relationship between CSRD of previous year and corporate market
value Tobin’Q of the following year (β = 0.044 and sig=0.00).
Model 2 with dependent variable TBQ

1

ROA

No

Yes

Yes (LEV)

2

TBQ

Yes

Yes

Yes(CSRD)

3

ROA

No

Yes

Yes (LEV)

4

TBQ

Yes

Yes

No

5

ROA

No

Yes

Yes (LEV)

6

TBQ

Yes

Yes

To overcome the endogenous phenomena in model 2, in this study, two-stage least

Yes (CSRD)

squares regression model (2SLS) was used to examine the relationship between corporate
social responsibility disclosure and corporate value. In this study, instrument variables can
be used for two-stage least squares regression 2SLS for the endogenous relationship
between corporate social responsibility disclosure and corporate value TBQ are audit
quality (AUDIT) and law (LAW). The first variable AUDIT means enterprises audited by
large auditing firms (Big 4) have a higher level of corporate social responsibility disclosure

than other enterprises. The second variable LAW means the more mandatory requirements

(Source: compilation of the author)

by the law, the higher the level of corporate social responsibility disclosures. The author
conducted regression analysis OLS, FEM, REM to verify the relationship between
instrument variables AUDIT, LAW and endogenous variable CSRD. The regression results

To overcome limitations of models, the author used different regression methods

showed that the selection of instrument variables is appropriate. This indicates that listed
companies audited by Big 4 auditing companies and the legal requirements increase result
in higher level of CSRD.

4.4.4. Regression results

depending on the type of problems
Model 1 and model 3 with dependent variable ROA:
The endogenous phenomenon exists in these two models as LEV is endogenous, in
this case, regression method GMM is used
Regression results of model 1 showed that implementation and disclosures of CSR
have impact on return on assets (ROA) of enterprises. In particular, CSRD immediately
affect ROA during the year with 0.0032% at a significance level of 1%. This result is
consistent with the hypothesis and studies of Mahoney and Roberts (2007); Mustaruddin
Saleh et al. (2011); Nagib Salem Bayoud et al. (2012); Dewi and Monalisa (2016), Elena
Platonova et al. (2016).
Regression results of model 3 showed that implementation and disclosures of CSR
in the past also have impact on return on assets (ROA) of enterprises in the following
year. Specifically, CSRD immediately affect ROA in the following year with 0.0032% at
a significance level of 1%. This result is consistent with the hypothesis and studies of

Ngoc Thao Trang and Liafisu Sina Yekini (2014), and Ho Thi Van Anh (2018) with the
same context of enterprises listed on the Vietnam stock market.
Model 4 with dependent variable TBQ
In order to overcome the phenomenon of heteroskedasticity and auto-correlation in the
research model, Generalized Least Square (GLS) is used. The regression results of GLS of

Next, the author conducted 2SLS (IV (2SLS) estimation) with dependent variable
TBQ, independent variable CSRD and instrument variables AUDIT, LAW. The
regression results showed that there exists a positive relationship between CSRD and
corporate market value Tobin'Q (β = 0.081 and sig = 0.03).
Model 5 with dependent variable ROA
The endogenous phenomenon exists in model 5 as LEV is endogenous similar to
model 1 and 3, thus regression method GMM is used to overcome this phenomenon
Regression results showed that CSRD has a positive impact on ROA of enterprises,
but the levels of impact of CSRD on ROA in the period of 2006 and 2007 are higher than
period since 2008 (P_value = 0.007 <0.05 and β = -0.00203. This result confirms the
research hypothesis that CSRD has a positive impact on financial performance (ROA)
and reinforcing the relationship between CSRD and financial performance of Mahoney
and Roberts (2007); Mustaruddin Saleh et al. (2011); Nagib Salem Bayoud et al. (2012);
Dewi and Monalisa (2016), Elena Platonova et al. (2016) and study of Ho Thi Van Anh
(2018)
Model 6 with variable TBQ
To overcome the endogenous phenomena in model 6 with independent variable
CSRD similar to model 2, two-stage least squares regression model (2SLS) was used to


17

18

examine the relationship between CSRD and corporate value TBQ. The regression results
showed that CSRD has a positive impact on corporate value TBQ (coefficient β = 0.168
and sig = 0.02), but impact on TBQ before 2008 is higher than that after 2008 (coefficient
β = -0.08898 and sig = 0.00). This results confirmed the research hypothesis that CSRD

The most disclosed topic in enterprises’ reports is “responsibility to employees”,
followed by “responsibility to customers, and “environmental responsibility” was
disclosed the least. This practice can be explained by stakeholder theory according to
Ullman's study (1985) that implementation and disclosures of CSR is to satisfy

has impact on financial performance with measure TBQ and consistent with studies of
Mustaruddin Saleh et al. (2011), Cahan et al. (2015), however, the level of impact during
the period from 2006 to 2007 is higher that during the period from 2008 to 2016

different interest groups of enterprises and parties with higher influential power would
receive more attention from enterprises and expectations of related parties has great
impact on enterprises.

indicating that the increase of stock prices depends on the conditions of the economy.
This result supplements signaling theory that CSR is a positive signal affecting investors'
decisions, but these decisions are also governed by other information, especially
information about the profitability of enterprises.

Thirdly, there are differences in CSRD between manufacturing and nonmanufacturing enterprises

CHAPTER5: DISCUSSION, RECOMMENDATION AND CONCLUSION

significant in the environmental indicator. It can be explained by the fact that
manufacturing enterprises have more impact on social environment than nonmanufacturing enterprises so it receives more attention and expectation from the social

5.1. Discussion
5.1.1. Discussion on establishing indicators of CSR
In this study, a list of indicators of CSR was established with four main groups of
corporate social responsibility: environmental responsibility, responsibility to employees,
responsibility to community, responsibility to customers in accordance with the
Vietnamese context. It is expected that the list of indicators of information can be applied
to further research on CSR due to its diversity and details of social responsibility issues,
easiness to understand and calculate. In addition, this list of indicators of CSR has a
significant contribution for policy makers, enterprise administrators, investors and social
community because it is a theoretical foundation for policies, actions associated with
social responsibility in Vietnam.

The survey showed that in manufacturing enterprises, the level of CSRD is higher
than that of non-manufacturing enterprises in all indicators of information, but most

community. It confirms finding of Deegan (2000) based on the stakeholder theory
providing that CSRD is related to expectations of stakeholders of enterprises.
5.1.3. Discussion on impact of CSRD on corporate finanical performance.
Table 5.1: Comparison of results and hypothesis
Model

Relationship

Expectation

Results

between
variables
1

CSRD - ROA

+

+

2

CSRD - TBQ

+

+

3

CSRD (-1) ROA

+

+

4

CSRD (-1) - TBQ

+

+

5

CSRD - ROA

+

+ (the level of impact after 2008 is lower)

6

CSRD - TBQ

+

+ (the level of impact after 2008 is lower)

5.1.2. Discussion on current situation of CSRD
Firstly, there is an increasing level of CSRD of companies listed on the Vietnam
stock market over time
The level of CSRD of listed companies on the Vietnam stock market has been
improved over time but still at a low level. Especially, in 2015 and 2016 there is a
remarkable change in the level of CSRD of enterprises. The results showed that pressures
from stakeholders such as government, community, and customers has motivated
enterprises to implement and disclose information related to rights and interests of
parties. This fact supports the stakeholder theory and legitimacy theory in explaining
CSRD is to meet the requirements of stakeholders to help businesses survive and
develop.
Secondly, there are differences in level of CSRD of each group of information

Source: Compilation of the author
The table above shows that CSRD positively affects both ROA and TBQ in the
current and future years. At the same time, there are differences in the impact of CSRD


19

20

on financial performance at different time periods. This result proved that
implementation of CSR during the course of operating business activities can create to
positive results to stakeholders.

time, these organizations and individuals need to take measures to protect their rights
from dishonest behaviors, unsafe products, threatened living and working environment.

5.1.4. Discussion on impact of law and audit quality on CSRD

5.3. Contributions of the research
5.3.1. Scientific and theoritical aspect

By testing appropriateness of the instrument variables, two variables were
identified: firstly, the audit quality variable (AUDIT) has a positive influence on
corporate social responsibility disclosures (CSRD); secondly, laws (LAW) have a

- Firstly, the thesis contributes partly and enriches theoretical foundation in the
field of accounting, especially on “information disclosure” in developing economies.
Besides, it contributes to improving the understanding of the stakeholder theory by

positive influence on CSRD of listed companies on Vietnam stock market. This result

shows that if the law is strengthened and the quality of auditing is high, the level of
CSRD would be high.

explaining that CSRD is associated with the relationship between enterprises and
related parties through analyzing the differences in the level of CSRD between two
types of enterprises: manufacturing enterprises and non-manufacturing enterprises.
At the same time, it strengthens the stakeholder theory, legitimacy theory and

5.2. Recommendations
5.2.1. Recommendations to the government

signaling theory by confirming the impact of CSRD on financial performance of
companies.

Firstly, the government should standardize requirements of CSRD
- Secondly, the research contributes to the understanding of CSRD of developing
Secondly, the government should gradually institutionalize CSRD in the
Vietnamese legal system into the Law on Enterprises as a mandatory responsibility.
Thirdly, the government should encourage enterprises to implement and disclose
CSR guaranteed by the third party.
Fourthly, there are measures to encourage implementation and disclosures of CSR
by strengthening the role of stakeholders.
5.2.2. Recommendations to listed companies
Firstly, raise awareness of enterprises on practice and disclosures of CSR
Secondly, take advantages of guidelines, consultation of organizations, experts
regarding necessary information or how to implement and disclose CSR.
Thirdly, complete accounting information system
5.2.3. Recommendations to investors
- Investors should have certain understanding of CSR, and consider the actual
practice of CSR in enterprises.

- A sign to recognize enterprises with good practice and disclosures of CSR is that
enterprises are audited by prestigious auditing firms.
5.2.4. Recommendations to other stakeholders of enterprises
Employees, customers, suppliers need to improve their understanding of CSR and
prioritize those companies who have good practice and disclosures of CSR. At the same

countries, especially Vietnam - a country with a slow growth of the economy compared
to other countries in the world by providing the list of detailed indicators of CSR
implemented and disclosed by companies listed in Vietnam stock market
- Thirdly, the thesis introduces and applies additional testing methods in the context
of research in Vietnam which are 2 SLS, GMM method in the model having endogenous
phenomenon with panel data to provide reliable evidence of the impact of CSRD on
financial performance of listed companies on the Vietnam Stock Market.
5.3.2. Practical aspect
- Firstly, the thesis shows the trend and reality of CSRD with four main groups:
Information on environmental responsibility, information on responsibility to employees,
information on responsibility to the community, information on responsibility to
customers of enterprises in general, and manufacturing and non-manufacturing
enterprises in particular listed on the Vietnam stock market during the period from 2006
to 2016.
- Secondly, the research provides evidence in Vietnam of the positive impact of
CSRD on enterprise financial performance with two representative indicators of return on
assets ROA and corporate value Tobin'Q. In addition, it pointed out differences in the
impact in different economic contexts with a sample of 473 observations of 43
enterprises over a long period of time from 2006 to 2016.


21

22

- Thirdly, the research provides evidence that enterprises listed on Vietnam stock
market with good auditing quality have high level of CSRD. In addition, this study also
shows that requirement of the law is an important factor positively affecting the level of
CSRD.

- The author hopes that in the future, it is possible to conduct research on impact of
CSRD on financial performance of enterprises with different levels of general
information disclosures.

Based on both theoretical and practical contributions, the author expects this
research to not only enrich the database for the field of research on practice and
disclosures of CSR in Vietnam but also provide meaningful reference for policy

The corporate social responsibility has becoming a global trend and an urgent
requirement for enterprises as it is considered a useful tool to help businesses
communicate their practice of social responsibility with stakeholders. This

makers, business executives and investors to enhance the level of CSRD of
Vietnamese enterprises for the purpose of sustainable development.

CONCLUSION

- Firstly, the content analysis method was used to measure the level of CSRD.

information contributes to improving the relationship between enterprises and related
parties and helps businesses achieve certain benefits. Although the main purpose of
this research is to examine the impact of CSR on financial performance of enterprises
but in the process of conducting research, the author has obtained some following

Although overcome subjectivity in the research through the cross-analysis between
assistants and author during the analysis process, certain errors and mistakes may exist
during the process. The content is based on the survey of annual reports and sustainability

results: Firstly, the study developed a detailed list of indicators of CSR that companies
listed in the Vietnam stock market often disclose with four information categories:
environmental responsibility, responsibility to employees, responsibility to the

reports of 57 enterprises with the largest and lowest market capitalization in the industry.
However, errors and subjectivenesss of the author are inevitable.

community and responsibility to customer. Secondly, it pointed out the situation and
differences in CSRD of two types of enterprises, namely manufacturing enterprises
and non-manufacturing enterprises. Thirdly, it showed a positive relationship between
CSRD and financial performance measured by the two indicators which were return

5.4. Limitations of the study

- Secondly, although there are many indicators to measure financial performance,
the author only chooses two representative criteria to measure the financial performance
of enterprises which are return on asset ROA and corporate market value Tobin'Q.
- Third, in this study, the author sought to identify impact of CSRD on financial
performance of enterprises with different scale, financial leverage (debt ratio) and different
growth rates, but have not considered different levels of general information disclosure of
each enterprise. So it is possible to miss important information that is not social
responsibility information but affects corporate financial performance.
5.5. Directions for future research
In this study, the author used the content analysis method to examine annual
reports and sustainability reports of enterprise based on the list of indicators that the
author built and adjusted according to the reference of previous studies. Therefore, in

the future when Vietnam's economy is growing, CSR is more standardized, the
measurement of CSRD will be easier and more reliable.
- In this study, the author used only two indicators to measure financial performance
which were return on assets ROA and corporate value Tobin'Q and selected a large
number of enterprises to ensure the integrity of the sample. The author hopes in the future
it is possible to use more indicators to measure financial performance and larger number
of enterprises to supplement research results.

on assets ROA and corporate market value Tobin'Q. Fourthly, it found a positive
effects relationship between CSRD and audit quality (AUDIT) and legal regulations
(LAW). With such research results, the author has also proposed some suggestions to
improve the level CSRD of companies listed on the Vietnam stock market. The author
believes that the results of the thesis are meaningful to management agencies,
businesses, investors and stakeholders of listed companies on the Vietnam Stock
Market.