Factors Affecting the Application of Environmental Accounting in Manufacturing Enterprises in Vietnam

Purpose: Starting from the goal of economic development associated with environmental protection activities, this study examines the factors affecting the application of environmental accounting (EA) in manufacturing enterprises in Vietnam. Methodology: This study is conducted using a combination of qualitative research methods (review of previous theories and research papers related to the research) and a quantitative research model (testing the degree of the appropriateness of the scale and the theoretical model through Cronbach's Alpha reliability coefficient, exploratory factor analysis (EFA), multivariable regression analysis to determine the relationship between factors affecting the application environmental accounting in manufacturing enterprises in Vietnam. Results: Multivariable regression analysis shows that all factors (six factors) included in the study have an impact on the application of environmental accounting in manufacturing enterprises in Vietnam. Which stakeholder pressure has the strongest impact and financial resources have the weakest effect on the application of EA in these enterprises. Originality/Value: This study provides empirical evidence about the impact of each factor on the application of environmental accounting, thereby helping state management agencies, professional associations, and businesses in the process of

Therefore, to combine socio-economic development with sustainable and long-term environmental protection in the future, enterprises in Vietnam need to implement environmental accounting (EA) in their operation and business management. In fact, EA is a useful tool to help business managers control costs more effectively; besides, it brings not only financial benefits but also brings great competitive advantages to enterprises in the current period of integration into the world economy. EA provides enterprises with the information they need to reduce their environmental impact, theory to examine the factors and barriers affecting the application of EMA in practice in small and medium enterprises in Malaysia, Jamil et al. (2015) find the factors that apply coercive force have a significant influence on the application of EMA, if the coercive pressure increases, SMEs are willing to implement the EMA. On the other hand, financial constraints are the biggest barrier to EMA development in SMEs in the manufacturing sector. In addition, the lack of knowledge and skills related to the environment and the absence of any guidelines on EMA have made it difficult to incorporate environmental issues into the accounting system. According to Alkisher (2013), education level affects the company's intention to use EMA. In other words, insufficient knowledge about environmental costs and benefits (Jamil et al., 2015) and a lack of skills will limit the integration of environmental issues into accounting systems in SMEs (Tran et al., 2021). Ofoegbu & Megbuluba (2016) examined the impact of factors belonging to the characteristics of manufacturing enterprises in Nigeria on the implementation of environmental accounting. The regression analysis results show that the business sector factor and the size of the enterprise have an impact on the implementation of EA in these enterprises. The institutional characteristics of the environment are increasingly being noticed as important determinants of organizational structure and performance (Hussain & Gunasekaran, 2002). However, businesses tend not to apply EMA when there is no pressure from state management agencies, so increasing coercive pressure, especially from the government will increase the intention to apply EMA (Chang, 2007). Government regulations and legal documents of environmental management agencies influence the disclosure of environmental information of companies (Nguyen et al., 2020); therefore, the development of the EMA will be handled by the Government and other authorities (Jamil et al., 2015). Pressure from the government, environmental protection organizations, and inter-agency environmental regulatory agencies significantly affect the ability to implement EMA practices (Tran et al., 2021). On the other hand, according to IFAC (2005), applying EMA will add value to the traditional management accounting system. At the same time, it provides valuable information for businesses in the process of management and improvement towards sustainable growth. In addition, the EMA is also seen as a support tool for external stakeholders interested in environmental performance. In addition, some studies show that financial resources play an important role in EMA adoption (e.g., Wachira, 2014;Jamil et al., 2015), and financial resources also have a significant influence on environmental disclosure activities (Nurul Huda, 2015), despite the increasing environmental awareness, the lack of financial resources is a major obstacle for SMEs to implement a environment management system in their business (Gadenne et al., 2009).

Isomorphic Institutional Theory
DiMaggio & Powell (1983) consider isomorphism as the concept that best describes the homogenization process. The authors believe that the organization's characteristics will be changed to match the features of the environment. The institutional theory of isomorphism deals with three elements: coercive isomorphism, simulated isomorphism, and normative isomorphism. From the point of view of coercive isomorphism, the legal system is the primary determinant of EA adoption. Enterprises may face the pressure of institutional pressure, and they must apply EA to satisfy the expectations and requirements of stakeholders, especially those with strong influence, such as regulators or owners; even coercive pressures come from customers, investors, competitors, etc.

Legitimacy Theory
The theory of legitimacy holds that an organization's activities must conform to the values or social norms in which it operates. The failure of organizations to adhere to social values or norms can make it difficult for the organization to gain the community's support to continue working. The sustainable development of enterprises must be considered from three perspectives: economic, environmental, and social (Elkington, 1997). Accordingly, an enterprise operates to create economic values while protecting and developing environmental and social values. However, the reality shows that enterprise activities always have increasingly serious impacts on the environment and social community. Therefore, the pressure of legal authorities and environmental groups as well as the social community forces businesses to implement environmental responsibility through environmental management according to standards and changing the accounting system to disclose appropriate environmental information. Pressure from society will force enterprises to fulfill their responsibility for the environment through standards and change the accounting system to disclose appropriate environmental information.

Firm Size
Firm size affects the structure and use of management controls in the enterprise. A larger enterprise typically has larger total resources and better internal information systems, facilitating EA adoption. When examining the factors affecting the application of EA in enterprises, Christ & Burritt (2013) and Ofoegbu & Megbuluba (2016) show that firm size affects EA implementation; precisely, the larger the enterprise's revenue, the more feasible the EA implementation will be. Based on the above argument, the first hypothesis of the study is stated as follows: H1: Firm size positively affects the application of EA in manufacturing enterprises.

The Pressure of Stakeholders
Stakeholders such as shareholders, creditors, consumers, employees, suppliers, or society, can be seen as people interested in social and environmental activities business. Stakeholders significantly impact the production and business activities of enterprises in the community. When considering the factors affecting the application of EA in enterprises, AlKisher (2013) and Jamil et al. (2015) show that pressures from the community or the media and the government agency on environmental protection standards will put pressure on enterprises to apply environmental accounting.
Through the above arguments, the second hypothesis of the study is stated as follows: H2: The pressure of stakeholders positively affects the application of EA in manufacturing enterprises.

Perception of the Benefits of Implementing EA
subjects towards EMA plays an important role in the application and development of EMA in practice. The issue of perceived usefulness reflects an understanding of the EMA platform, the role and importance of EMAs in helping corporate governance, and connecting the interests of businesses with social benefits. Applying EMA helps improve the organization's image but also increases the relationship with the community and stakeholders, complies with environmental laws, avoids fines and compensation, and fixes environmental problems (IFAC, 2005). Burritt et al. (2002) and Ferreira et al. (2010) mention the benefits of EMA and argue that EMA helps businesses identify cost-saving opportunities and make decisions about improving product structure and pricing to avoid future costs associated with investment decisions and improve financial performance. Through the above arguments, the third hypothesis of the study is stated as follows: H3: Perception of the benefits of implementing EA positively impacts the application of EA in manufacturing enterprises.

Legal Regulations
Institutional features of the environment are increasingly being noticed as essential determinants of organizational structure and performance (Hussain & Gunasekaran, 2002). Coercive factors significantly impact EMA performance (Jamil et al., 2015).
Therefore, without government pressure on organizations to implement EMA, organizations are less likely to voluntarily apply EMA (Chang, 2007). The lack of EMA guidance is also a barrier to integrating environmental issues into existing accounting systems and practices, and the lack of EMA guidance makes it difficult to collect, identify, analyze and effectively evaluate data on the environment, especially waste management, prevention of environmental pollution (Jamil et al., 2015). Based on the above arguments, the fourth hypothesis of the study is stated as follows: H4: Legal regulations directly and positively influence the application of EA in manufacturing enterprises.

Financial Resources
Cost is an essential factor in implementing any new system. This is one of the issues that many businesses are concerned about. Although the benefits of applying EMA are great, the cost of organizing the EMA system is not small. Therefore, finances are the most critical barrier preventing businesses from implementing EMA. Financial resources are one of the factors affecting the level of EMA adoption (Wachira, 2014; Jamil et al., 2015) as well as environmental disclosure (Nurul Huda, 2015); despite the increasing environmental awareness, the lack of financial resources is a significant obstacle for SMEs to implement environmental management systems in their enterprises (Gadenne et al., 2009). Based on the above arguments, the fifth hypothesis of the study is stated as follows: H5: The larger the financial resources of the enterprise, the higher the possibility of applying EA in manufacturing enterprises.

Staff Qualifications
Staff qualifications are considered an important factor in successfully applying to EA because if the employees do not master the knowledge and skills, the application of EA in the enterprise will face difficulties. Alkisher (2013) showed that education level is one-factor affecting enterprises' intention to apply EMA. Applying EMA tools is considered too complicated and requires highly qualified staff (Tran et al., 2021); therefore, a lack of knowledge and skills also limits the integration of environmental issues into the accounting system in SMEs (Jamil et al., 2015). Based on the above principles, the sixth hypothesis of the study is stated as follows: H6: Staff qualifications impact and influence in the same direction as the application of EA in manufacturing enterprises.
Based on the above hypotheses, the authors propose a research model presented as shown in Figure 1.

Data Collection Methods
The data of the study were collected through primary and secondary data sources.
Secondary data is collected from research-related sources such as domestic and foreign journals and articles that have been published. Primary data was collected through a survey using a questionnaire with a 5-point Likert scale (in which 1-Strongly disagree, 2 -Disagree, 3 -Normal, 4 -Agree, 5 -Strongly agree) by

Sampling Method
Sample size: in multiple regression analysis, the sample size is calculated by the formula n >= 50 +8p (p -the number of independent variables in the model, n -sample size) (Tabachnick & Fidell, 2007). According to Hair et al. (1998) Survey sample characteristics: the authors sent 300 questionnaires to the surveyed subjects at manufacturing enterprises in Vietnam. As a result, the number of valid questionnaires is 265 (an 88.33 percent response rate). The remaining 35 questionnaires were invalid due to multiple-choice a question, leaving the required answers blank. Thus, the sample size used in the study is 265 (the sample size is completely suitable in the multiple regression analysis). In this study, the characteristics of the study sample are shown in Table 1.

Data Analysis Method
After data cleaning, the author uses SPSS 20 software for analysis, and data processing and runs multiple linear regressions through criteria such as testing the reliability of Cronbach's Alpha scale, exploratory factor analysis (EFA), and multivariable linear regression test to consider the impact of factors on the application of EA in manufacturing enterprises in Vietnam.

Scale Reliability Test Results
This study uses Cronbach's Alpha test to test the close correlation between the observed variables of a factor included in the regression model. The results of the reliability test of the scale presented in Table 2 show that all variables included in the regression model have Cronbach's Alpha coefficients greater than 0.6, so the scale of the variables is both reliable and usable. The reliability test results of the scale presented in Table 2 show that all variables included in the regression model have Cronbach's Alpha coefficients greater than 0.6, so the scale of the variables is both reliable and usable (Hoang & Chu, 2008). Thus, through testing the scale's reliability, the observed variables of the factor groups remain the same.

Analysis of the Scale of Factors Affecting the Application of EA
The test results presented in Table 3, Table 4, and Table 5 show that the KMO coefficient is 0.738 (satisfying the condition 0.5 ≤KMO ≤ 1), so it is satisfactory; the bartlett's test has a Sig value of 0.000 (less than 0.05), so these observed variables have a close relationship with each other and are suitable for EFA analysis. The total variance extracted is 69.407% (satisfying the condition greater than 50%) at the 1988). The observed variables in the group of factors all have factor loading coefficients greater than 0.5, satisfying the requirements (Hair et al., 1998). Thus, after conducting EFA analysis, the extracted factors are reliable and valid.

Analysis of the Applicable Scale EA
The test results in Table 6 and Table 7 show that the KMO coefficient is 0.860 (satisfying the condition 0.5 ≤KMO ≤ 1), and the Sig value of the bartlett's test is 0.000 (less than 0.05), so the observed variables are correlated in the population. Therefore, the total variance extracted is 60.681% (greater than 50%); at Eigenvalues of 3,641 (greater than 1), the model is eligible to conduct EFA analysis.

Source: Research data analysis results
The results of Table 8 show that the observed variables of the dependent variable MT all satisfy the factor loading greater than 0.5. Therefore, after EFA analysis, the six observed variables of the dependent variable MT (from MT1 to MT6) remain the same.

Multiple Linear Regression Analysis
Spearman's correlation test tests the relationship between independent and dependent variables. The results in Table 9 show that the independent variables QM, AL, LI, PL, TC, TD, and the dependent variable MT all have Sig less than 5%, so these independent variables are correlated with the dependent variable and will be included in the model to account for the dependent variable.

Source: Research data analysis results
This study uses multivariable regression analysis to test the research hypotheses by examining the relationship between the independent variables and the dependent variable. The results of multivariable regression analysis are presented in Table 10, and Table 11 shows that the Adjusted R Square value is 65.8%, which means that the independent variables explain 65.8% of the variation of the dependent variable. In addition, the ANOVA test results have a Sig value of 0.000 (less than 0.05), so the regression model is suitable and statistically significant. This shows that at least one independent variable in the regression model affects the dependent variable.

Source: Research data analysis results
The results of multivariable regression analysis in Table 12 show that the variables QM, AL, LI, PL, TC, and TD all have Sig values less than 0.05, so the regression model is statistically significant and suitable with the data set included in the study, i.e.
independent variables QM, AL, LI, PL, TC, and TD have an impact on the dependent variable MT. The variance inflation factor is less than 2.20, so multicollinearity does not occur (Nguyen, 2011).

Source: Research data analysis results
In general, after testing the research hypotheses by multiple linear regression model,

Firm Size
The regression results show that firm size is an important factor, positively affecting the application of EA in manufacturing enterprises (B1 = 0.285). This finding proves that when the firm's size increases by 1 unit, the application of EA in the enterprise increases by 0.285 units. This implies that the larger the enterprises, the higher the likelihood of EA adoption. The results of this study are consistent with the findings of Christ & Burritt (2013), and Ofoegbu & Megbuluba (2016). When the firm size is not large, business managers often give priority to the performance of accounting work to provide accounting and economic information related to financial accounting to meet the needs of customers compliance with the regulations of the authorities (especially the tax authorities) rather than the transparency of the data, the information provided, including environmental information; therefore the application of EA at these enterprises is also of little interest.

The Pressure of Stakeholders
The research results show that stakeholders' pressure is an important factor, having the strongest effect on the application of EA in manufacturing enterprises (B2 = 0.346). This result proves that if stakeholders' pressure increases by 1 unit, the application of EA in enterprises increases by 0.346 units. In other words, the greater the stakeholder's pressure, the higher the application of EA in enterprises. This finding is entirely consistent with the research results of AlKisher (2013), Jamil et al. (2015) and Tran et al. (2021). In Vietnam, many guidelines and policies on environmental protection and response to climate change are being paid special attention by the government; the legal system on the environment is increasingly perfected with the goal of protecting the environment. Agencies often require enterprises to report and assess their environmental and social impacts when preparing their annual reports, which forces enterprises to have systems in place to manage financial, environmental, and performance society.

Perception of the Benefits of Implementing EA
The research results show that the perception of the benefits of implementing EA positively affects the application of EA in manufacturing enterprises (B3 = 0.195 Burritt et al. (2002) and Ferreira et al. (2010). However, Vietnam does not have many manuals as well as indepth teaching programs on EA, so it is difficult for enterprises to monitor, record and analyze environmental information. Therefore, the perception of the benefits of implementing EA in the business community still has certain limitations.

Legal Regulations
The research results show that legal regulations play an important role and positively impact the application of EA in manufacturing enterprises (B4 = 0.242). In other words, when legal regulations increase by 1 unit, the adoption of EA in enterprises increases by 0.242 units. This indicates that the lack of mandatory regulations or implementation guidelines will make it difficult to apply EA in enterprises. This issue also shows that EA depends greatly on whether the authorities have issued legal documents related to EA. Thus, it can be affirmed that to promote the application of EA in enterprises; the most important thing is the element of "legal regulations" only when there are regulations under pressure will enterprises perform. The findings of this study are consistent with those of Chang (2007), Jamil et al. (2015) and Nguyen et

Financial Resources
Financial resources continue to be a factor that positively impacts the application of EA in manufacturing enterprises (B5 = 0.101). Research results imply that when financial resources increase by 1 unit, the application of EA in enterprises increases by 0.101 units. Normally, to apply EA, enterprises need to invest in infrastructure for EA application, such as information technology systems, information processing processes, etc., and businesses must train human resources teams with professional qualifications and skills in handling information related to EA. Therefore, to apply EA, enterprises need to invest a large amount of money to have a good infrastructure system and the expenses to train staff to serve the implementation of EA. Thus, it can be seen that financial resources are an obstacle to the application of EA in enterprises; enterprises with restricted financial resources will find it tricky to apply EA. This result is consistent with the study of Gadenne et al. (2009), Wachira (2014), Jamil et al. (2015) and Nurul Huda (2015).

Staff Qualifications
The research results show that staff qualifications positively impact the application of EA in manufacturing enterprises (B6 = 0.139). This shows that when the staff qualifications increase by 1 unit, the ability to apply EA in enterprises increases by 0.139 units. Indeed, the team of accountants in an enterprise greatly influences the organization of accounting work; it affects the process of receiving and processing accounting information to provide managers with accurate and timely decisionmaking. Therefore, enterprises with a team of trained, qualified, knowledgeable, and experienced accountants will help apply EA more smoothly. This result is similar to that of Alkisher (2013), Jamil et al. (2015) and Tran et al. (2021). However, in Vietnam, most accounting training programs have not focused on EA training, so many accountants in enterprises lack knowledge and expertise in EA, leading to the application of EA in Vietnamese enterprises being limited.

Conclusions
As a developing country, Vietnam is strongly affected by climate change, pollution, and environmental degradation, which has affected the economy as well as people's quality of life. The main object of environmental degradation is enterprises; most enterprises in developing countries often put profit first instead of growth associated with environmental protection activities. Therefore, starting from the goal of propaganda to raise awareness and qualifications of managers in enterprises about system on this content. For businesses, it is necessary to change awareness and social responsibility for environmental issues; and develop a long-term business strategy that considers the impacts of product environmental standards and regulations. At the same time, business administrators must regularly update and implement EA.
Finally, training institutions must continue to supplement, edit and revise to improve the quality of teaching curricula on EA; put EA into teaching as an intensive course, organized into topics for students to exchange, discuss and draw experience.
Although specific results have been achieved, this study still has some limitations that future studies need to consider and expand the research. First, the data used in the study is collected through surveys in many types of enterprises (different sizes, industries, etc.), so it is not representative of each type of business. Secondly, many other factors can affect the application of EA, such as environmental strategy, business lines, audit, etc. but have not been considered in this study. Finally, the study used analytical and synthesis methods from published studies for a long time, and the sample size was small, so the generalizability was not high, affecting the quality of the study.