Environmental Dimension of Corporate Social Responsibility and Earnings Persistence: An Exploration of the Moderator Roles of Operating Efficiency and Financing Cost

Abstract

Society has gradually realized that companies’ actions have consequences. Companies can fulfill their accountability to society by disclosing information beyond their financial data, providing better decision making for stakeholders. Therefore, this study aims to investigate the impact of corporate social responsibility (CSR) on earnings persistence (EP) for the moderator roles of operational efficiency and financing cost for the companies listed on the Tehran Stock Exchange. For this purpose, the statistical population consists of 714 firm-year observations from 2014 to 2020 (7 years). A multivariate regression method was used based on the panel data analysis method to test the research hypotheses. The results indicate that corporate social responsibility for earnings persistence has a positive and significant relationship with the moderator role of operational efficiency, but is unrelated to the moderator role of financing cost. The majority of the prior research in this field has focused on developing countries. An international perspective is critical, and this study helps draw a more contextualized picture of sustainability than before.

1. Introduction

Countries worldwide are reinforcing regulations to encourage firms to establish biocompatible technologies in response to the worsening environmental conditions of the world [1]. Policies, such as taxes on environmental damage, can be adopted for this purpose. These policies can target economic enterprises that operate as the primary sectors of the market and account for the major portion of pollution emissions [2].

Hart [3] believed businesses operate in a world with increasingly narrowing resources, meaning their competitive advantage will lie in activities which revitalize the environment. Biocompatible activities can thus be considered job opportunities that will determine firms’ future visions for growth [4]. Environmental investments can protect firms against future auditing costs and reduce business uncertainty [5]. Firms that fulfill their environmental duties may take advantage of the payoffs in the long run, despite the costs they may have to endure over the short term [6].

Regarding creating value, ref. [7] inferred that reducing environmental risks increases productivity. Flack and Heblich [8] found that environmental responsibility could enhance employee efficiency, reduce a business unit’s operational costs, and improve a firm’s reputation.

In other words, firms can improve their reputation through activities associated with social responsibility. Meanwhile, stakeholders can play a prominent role in improving a firm’s reputation by holding it to the social responsibility standard [9]. Such firms seek to avert incidents with negative impacts, disclose information, and hold themselves accountable to neutralize any negative consequences [10]. Firms that operate more transparently regarding the operations that are associated with the community and environment are found to be more financially stable and to suffer fewer fluctuations [11]. Moreover, the quality of earnings in these firms is more desirable than that of others [12].

The persistence of earnings is an index of earnings quality. An earning is considered persistent when it helps stakeholders assess the stability of businesses and predict their future performance [13]. Penman and Zhang [14] and Richardson et al. [15] believed that participants in the stock market use earnings persistence to evaluate stock values, since earnings hold important information for a firm’s performance assessment. The persistence of companies in creating profit is an essential aspect of performance assessment. Park and Shin [16] argued that an earning is persistent as long as it is stable, permanent, and recurrent.

In this regard, ref. [17] suggested that firms adhering to social responsibility are less involved in earnings management activities and will perform more transparently. Therefore, according to the stakeholder theory, more accurate financial reports are expected to result in high-quality earnings. Prior et al. [18] believed that managers may want to divert the attention of stakeholders from their opportunistic operations by taking measures associated with social responsibility. Relying on the agent theory, the assumption is raised that managers have different incentives from shareholders and seek to maximize their advantage over the short term. Čajka et al. (2023) believed that having an executive director from outside the company could be very beneficial for companies [19]. A manager needs to possess specific management skills and be invested in the company’s prosperity [20]. Managers are responsible for achieving organizational objectives through versatile prowess and managerial knowledge, coupled with excellent decision-making skills [21]. Capable managers who make the right decisions can ultimately improve the value of a firm [22].

Relevant studies indicate that the relationship between CSR and earnings quality can be positive or negative. Firms send a specific signal to their stakeholders when they prove committed to CSR activities [23]. There is a chance of firms using CSR activities to manipulate their stakeholders with the excuse of responding to their demands. Earnings quality thus appears vital to gain the trust of stakeholders and, consequently, meeting the demands of shareholders and creditors [24].

Furthermore, the signaling theory indicates to what extent firms send the necessary signals regarding their private information to financial statement users [25]. The persistence of earnings is a signal explaining earnings stability in the future. This index helps creditors and investors use earnings information as a consideration in their economic decisions. Turning to environmental responsibilities, the signaling theory explains that firms seek to send a signal regarding their quality of information [26,27]. This could signal high information quality [28] or persistence [29].

Meanwhile, information risk declines with information transparency. As a result, firms operating actively in terms of their environmental responsibilities pay more attention to the stakeholders, and they are seen as more trustworthy from the investors’ and stakeholders’ viewpoints, and can access financing institutions more easily [8]. Therefore, the theories mentioned above indicate that the more a firm is committed to its social responsibilities, the more it will gain the investors’ trust [30]. Since CSR will increase and improve the reputation of companies [31,32], it can be said that it will improve the financial performance of companies [33,34,35,36]. Also, it will allow foreign lenders to give the necessary support to companies so that they can invest in new and green technologies [37,38]. Finally, the operational and financial performance of the company should be improved.

CSR is a complex concept that has been the subject of notable academic studies in accounting globally. It is related to the ethical and moral aspects of corporate decision making and behavior, and it addresses various issues, such as environmental protection, the management of human resources, health and safety at work, community relations, and relationships with suppliers and customers [39]. Engagement in socially responsible activities enhances stakeholders’ satisfaction, positively impacts corporate reputation [40], and mitigates financial risk [41]; therefore, it can significantly affect the performance of companies. This will improve the profitability of companies. Hence, the primary purpose of this research is to investigate the relationship between CSR and earnings persistence. In addition, we intend to examine the moderating roles of operational efficiency and financing cost variables in this regard.

This study has practical implications. Since investment in the social and environmental responsibility of companies is accelerating in the market, the persistence of earnings can be effective in companies’ value creation and performance. Therefore, this study’s findings can be considered useful for investors and creditors interested in the issue of corporate sustainability. This study contributes to the literature in at least three aspects. First, the present research investigates CSR’s influence on earnings persistence in emerging markets, such as Iran. The resulting experimental evidence can benefit policymakers in developing countries in compiling CSR policies and regulations. Second, the present study performs an additional test to examine whether CSR affected the earnings response coefficient. Third, this study examines corporate social responsibility mechanisms, such as financing costs and operating efficiency, that have an influence on economic consequences using a model incorporating the moderating effects. It is worth noting that in the present study, we consider the environmental dimension of CSR.

2. Research Background and Hypothesis Development

The disclosure of financial and non-financial information about a company’s impact on the environment and society is known as CSR reporting. Carroll [42] argues that CSR encompasses a society’s economic, ethical, legal, and political expectations. In addition, paying attention to the legal system of countries can be very important [43]. It should be noted that an analysis of the behavioral standards of investors from the perspective of a reference to regulatory laws (such as regulatory independence, the political environment, and the flexibility to introduce new regulations) or a reference to social investment issues (such as human rights, labor, health, social responsibility, and poverty reduction) also plays a significant role [44]. Specifically, CSR reports on the economic, social, and environmental concerns affecting business are collected and published annually in corporate annual reports [45].

Stakeholder theory suggests that CSR facilitates effective stakeholder management and enables companies to create, develop, and maintain relationships with critical resources. Companies with superior CSR engage in more stakeholder communication [45].

Signaling theory suggests that stakeholders rely on information to make decisions that serve their best interests [46]. A company’s competent CSR disclosure conveys a more positive image to the public than a company’s insignificant CSR disclosure [47,48]. Refs. [40,49] have shown that CSR disclosure can help reduce corporate risk. Building strong relationships with consumers and suppliers, and developing a reputation for high-quality products and services enables businesses to share socially responsible information. Furthermore, attention to environmental and social changes can help organizations to increase their performance, reduce the chances of earnings mismanagement, and thus maintain the stability of high-quality financial statements while increasing earnings persistence [45].

The role of business units in the community has recently undergone immense changes, to the point that business units are expected to increase their earnings and hold themselves accountable to their community [50]. A mutual relationship has thus emerged between the business unit and the community [51]. Firm responsibility can benefit the business unit itself and the community it resides in. A better understanding of its potential advantages can lead to superior investment returns for a firm [51].

Empirically, ref. [52] found that customers expect business units to observe a code of ethics in their administration. The authors of [53] also demonstrated that customers prefer to purchase from firms that are more aware of their social responsibility. Among the other advantages of firm that are socially responsible, one could point to the increased capacity to recruit new and retain experienced staff. Some of these employees help develop and operate communication systems, while others are responsible for supporting and improving the communication system or work in the area of data protection [54]. As they progress toward full social responsibility, firms realize that it is easier to recruit new forces and retain experienced staff for a sufficient time period to ensure the business unit’s success [55]. Government support is among the other advantages of being socially responsible, as altruistic businesses are more committed to social responsibility and are subject to fewer lawsuits from regulators, be it tax auditors, the government [51], or customers (company representatives, external auditors, competitors, and investors) [56].

Investors rely more on earnings information than other performance indices [57] (e.g., cash flows, cash earnings, and earnings changes). Reported earnings are one of the determining criteria for investors’ expected returns [58]. The presentation of information must be made in a manner that enables the assessment of past performance so that the presented earnings can help users to evaluate a firm’s performance and profitability, and enable investors to estimate their expected returns based on this information [59,60]. Aside from the fact that the reported earnings figure is important for investors and influences their decisions, the qualitative features of earnings can be of significance to the investors as one of the aspects of earnings information [61]. The persistence of earnings is among the qualitative features of accounting profits based on accounting information [62]. The persistence of earnings is an index which helps financial statement users and investors to assess a firm’s cash flows and future earnings, and refers to the persistence and recurrence of current earnings [45]. The higher the persistence of earnings, the more capable a firm will be at maintaining its current earnings. Assuming that the market influences CSR information, one could expect stock price changes to be associated with more unexpected profits for firms with higher degrees of social responsibility [63].

According to [64], the persistence of earnings is a crucial scale for the long-term performance of firms that report earnings information. The persistence of earnings is known as a suitable scale for firm performance. Fulfilling CSR helps to improve production technology, increase employee productivity, and establish a significantly higher operational efficiency in a firm [65]. The benefits of CSR include reputation enhancement, shareholder wealth improvement, better risk management, the improvement of market demand from customers, an increase in disclosure and reporting transparency, and an overall ability to access financial markets with better conditions [66].

Furthermore, adopting corporate social responsibility (CSR) initiatives can enhance internal communication and address human resource challenges, ultimately leading to economic efficiency [67]. To implement CSR, companies need to change the existing production mode and encourage the adoption of more environmentally friendly technologies. Using new and green technologies improves a system’s and a company’s overall productivity [68].

In addition, higher operational efficiency is a source of competitive advantage that will ultimately influence a firm’s performance. Corporate social responsibility is specifically dependent on higher-level duties that require changes to the current state of production, and incentives for firms to adopt biocompatible technologies [69]. Moreover, CSR can help establish a more active interaction between a firm and its employees, contributing to labor force productivity [70]. From a supply chain perspective, firms take on environmental responsibilities based on the needs of their stakeholders, such as customers, suppliers, and rivals, seeking to improve their reputation and attract consumers [71].

Empirical research has attempted to establish and explain the relationship between CSR and corporate financial performance. However, the results have been contradictory [72,73,74].

Therefore, companies are trying to use new and clean technologies in order to grow and improve their performance. As stated, CSR will raise the reputation of companies. This will allow companies to benefit from external support, such as from financial institutions and low-cost loans, in order to increase the financing capacity of companies’ debt, reputation, and transparency of information [75]. In other words, CSR allows companies to secure more debt financing, which can help mitigate operational uncertainty, reduce earnings volatility due to market changes, and enhance earnings persistence [2].

CSR helps polluting firms take advantage of external aid, such as low-interest loans and from financial institutions, to enhance their cost financing capacity, improve their reputation, and increase transparency [76]. In other words, CSR allows economic enterprises to gain better financial support for their liabilities, helping them fight operational uncertainty. Such firms can thus reduce the revenue uncertainty in the market resulting from fluctuations in the credit market and improve their persistence of income. Corporate social responsibility enables firms to take advantage of sustainable competitive advantages, such as improved relations between firms and creditors, to reduce default debt risk [2].

Hypothesis 1 (H1).

There is a relationship between the fulfilling of CSR by firms and the persistence of earnings with the moderating role of operational efficiency.

Hypothesis 2 (H2).

There is a relationship between the fulfilling of CSR by firms and the persistence of earnings with the moderating role of financing cost.

3. Research Methodology

3.1. Research Population

The statistical population of the present study encompassed the firms listed on the Tehran Stock Exchange, and the temporal scope of this study included the years 2015–2021 (seven years). Moreover, the total of 102 firms that were selected from those listed on the Tehran Stock Exchange were subjected to the following inclusion and exclusion criteria:

• The desired firms must have been listed on the Tehran Stock Exchange throughout 2015–2021.
• The financial year must be the same across all the firms, starting with the Solar New Year, to allow comparability.
• The firms must not have changed their financial period over the temporal domain of this study.
• The firms must not have changed their field of activity over the studied period.
• The firms must not be among the investment and mediation businesses, such as banks and insurance companies. The final sample consisted of 714 firm-year observations.

3.2. Research Model and Variables

Following [2], the basic Equation (1) was used to test the relationship between fulfilling CSR and the persistence of earnings.

$${\mathrm{ROA}}_{\mathrm{it}+1}={\mathsf{\beta }}_0+\mathsf{\rho }{\mathrm{ROA}}_{\mathrm{it}}+{\mathsf{\beta }}_1{\mathrm{POLLUTING}}_{\mathrm{it}}\times {\mathrm{ROA}}_{\mathrm{it}}+{\mathsf{\beta }}_2{\mathrm{CSR}}_{\mathrm{it}}\times {\mathrm{POLLUTING}}_{\mathrm{it}}\times {\mathrm{ROA}}_{\mathrm{it}}+\mathsf{\beta }{\mathrm{X}}_{\mathrm{it}}+\mathsf{\epsilon }$$

(1)

Equations (2) and (3) were used to test whether fulfilling CSR increased the persistence of earnings through an increase in operational efficiency and a decline in financing costs (to test the first and second hypotheses).

$${\mathrm{MEDIAT}}_{\mathrm{it}+1}=\complement +{\mathrm{m}}_1{\mathrm{POLLUTING}}_{\mathrm{it}}+{\mathrm{m}}_2{\mathrm{CSR}}_{\mathrm{it}}\times {\mathrm{POLLUTING}}_{\mathrm{it}}+{\mathrm{m}}_3{\mathrm{CSR}}_{\mathrm{it}}+\mathsf{\beta }{\mathrm{X}}_{\mathrm{it}}+\mathsf{\epsilon }$$

(2)

$${\mathrm{ROA}}_{\mathrm{it}+1}=\complement +\mathsf{\rho }{\mathrm{ROA}}_{\mathrm{it}}+{\mathsf{\beta }}_1{\mathrm{MEDIAT}}_{\mathrm{it}}\times {\mathrm{ROA}}_{\mathrm{it}}+{\mathsf{\alpha }}_2{\mathrm{CSR}}_{\mathrm{it}}\times {\mathrm{POLLUTING}}_{\mathrm{it}}+\mathsf{\beta }{\mathrm{X}}_{\mathrm{it}}+\mathsf{\epsilon }$$

(3)

MEDIAT is the moderating variable through which CSR influences the persistence of earnings in firms, and includes the operation efficiency rate (TURN) and financing costs (FC) indices.

The persistence of earnings in industries with firms that pollute the environment were estimated by entering the results of Equation (2) into Equation (3). The persistence of earnings will improve for the firms in polluting industries with increased operational efficiency and reduced financing costs if ${\mathsf{\beta }}_1$ and ${\mathrm{m}}_2$ are both significant.

X is a vector of the control variables, which will be later introduced in the operational definitions (Table 1).

Ohlson (1995) believed that the higher the earnings persistence, the more sensitive the investors’ reaction to earnings will be. As a result, earnings persistence is an important factor for investors, as it makes earnings an effective factor in stock pricing. Therefore, in the current research, various criteria have been used for a better test; for example, ROA, which is an index of accounting earnings, and MV, which is a stock price market index [77].

Table 1. Definition of research variables.

No.	Title	Symbol	Definition
Dependent variable
1	Book-to-market value	MV	Net assets to stock market value at the end of the current year.
3	Return on assets	ROA	Net profits to total assets at the end of the current year.
Independent variable
4	Corporate social responsibility	CSR	In the current research, the dimensions of environmental responsibility (such as biodiversity, pollution, energy consumption, etc.) are measured based on the disclosure of each of the following items in the activity report of the board of directors [78]: The policies, goals, and impacts of environmental protection. The firm’s annual energy consumption. The investment in environmental protection through advanced environmental technology equipment. The types, amounts, and concentrations of pollutants and their discharge locations. The environmental protection facilities and their construction. The purification and disposal of the waste generated in the production process. The voluntary environmental protection agreements signed. The rewards for environmental protection. The other environmental information disclosed voluntarily by the firm. The disclosure of each item earns a score of 1. The sum of the score is eventually considered as the environmental index.
5	Polluting industries	POLLUTING	A score of 1 if the firm is in a polluting industry (automotive, petrochemical, chemical, or pharmaceutical) and 0 otherwise.
Moderating variables
6	Financing costs	FC	The financial costs from all the institutions that were received.
7	Operational efficiency	TURN	The operational income to total assets.
Control variables
8	Accounting information quality	ABSDA	${\mathrm{DA}}_{\mathrm{it}}={\mathsf{\beta }}_0+{\mathsf{\beta }}_2\frac{1}{{\mathrm{TA}}_{\mathrm{it}}}+{\mathsf{\beta }}_2\frac{∆{\mathrm{Sale}}_{\mathrm{it}}}{{\mathrm{TA}}_{\mathrm{it}}}+{\mathsf{\beta }}_3\frac{{\mathrm{PPE}}_{\mathrm{it}}}{{\mathrm{TA}}_{\mathrm{it}}}+{\mathsf{\beta }}_4\frac{{\mathrm{CFO}}_{\mathrm{it}}}{{\mathrm{TA}}_{\mathrm{it}}}+{\mathsf{\epsilon }}_{\mathrm{it}}$ DA: accruals. TA: total assets. SALE: Sale. PPE: total property and machinery. CFO: operational cash flow. Accounting information quality index: the residual absolute value of Modified Jones.
9	Firm age	AGE	The logarithm of the number of years the firm has been in the stock market.
10	Board size	BOARD	The logarithm of the number of board members.
11	Environmental responsibility disclosure	DISCER	A score of 1 if the environmental responsibility score is higher than the median of the studied sample and 0 otherwise.
12	Financial leverage	LEV	Total liabilities to total assets.
13	Fixed assets	PPE	Fixed assets to total assets.
14	Firm size	SIZE	The logarithm of total assets.
15	Ownership state	STATE	A score of 1 if the shareholders are governmental bodies and 0 otherwise.

Table 1. Definition of research variables.

No.	Title	Symbol	Definition
Dependent variable
1	Book-to-market value	MV	Net assets to stock market value at the end of the current year.
3	Return on assets	ROA	Net profits to total assets at the end of the current year.
Independent variable
4	Corporate social responsibility	CSR	In the current research, the dimensions of environmental responsibility (such as biodiversity, pollution, energy consumption, etc.) are measured based on the disclosure of each of the following items in the activity report of the board of directors [78]: The policies, goals, and impacts of environmental protection. The firm’s annual energy consumption. The investment in environmental protection through advanced environmental technology equipment. The types, amounts, and concentrations of pollutants and their discharge locations. The environmental protection facilities and their construction. The purification and disposal of the waste generated in the production process. The voluntary environmental protection agreements signed. The rewards for environmental protection. The other environmental information disclosed voluntarily by the firm. The disclosure of each item earns a score of 1. The sum of the score is eventually considered as the environmental index.
5	Polluting industries	POLLUTING	A score of 1 if the firm is in a polluting industry (automotive, petrochemical, chemical, or pharmaceutical) and 0 otherwise.
Moderating variables
6	Financing costs	FC	The financial costs from all the institutions that were received.
7	Operational efficiency	TURN	The operational income to total assets.
Control variables
8	Accounting information quality	ABSDA	${\mathrm{DA}}_{\mathrm{it}}={\mathsf{\beta }}_0+{\mathsf{\beta }}_2\frac{1}{{\mathrm{TA}}_{\mathrm{it}}}+{\mathsf{\beta }}_2\frac{∆{\mathrm{Sale}}_{\mathrm{it}}}{{\mathrm{TA}}_{\mathrm{it}}}+{\mathsf{\beta }}_3\frac{{\mathrm{PPE}}_{\mathrm{it}}}{{\mathrm{TA}}_{\mathrm{it}}}+{\mathsf{\beta }}_4\frac{{\mathrm{CFO}}_{\mathrm{it}}}{{\mathrm{TA}}_{\mathrm{it}}}+{\mathsf{\epsilon }}_{\mathrm{it}}$ DA: accruals. TA: total assets. SALE: Sale. PPE: total property and machinery. CFO: operational cash flow. Accounting information quality index: the residual absolute value of Modified Jones.
9	Firm age	AGE	The logarithm of the number of years the firm has been in the stock market.
10	Board size	BOARD	The logarithm of the number of board members.
11	Environmental responsibility disclosure	DISCER	A score of 1 if the environmental responsibility score is higher than the median of the studied sample and 0 otherwise.
12	Financial leverage	LEV	Total liabilities to total assets.
13	Fixed assets	PPE	Fixed assets to total assets.
14	Firm size	SIZE	The logarithm of total assets.
15	Ownership state	STATE	A score of 1 if the shareholders are governmental bodies and 0 otherwise.

4. Results

4.1. Descriptive Statistic Results

Table 2. Results of descriptive statistics.

Variable	Symbol	Mean	Median	Maximum	Minimum	S.D
Book-to-market value	MV	0.935	0.685	6.311	0.007	0.935
Return on assets	ROA	0.112	0.102	0.626	−0.605	0.151
Corporate social responsibility	CSR	4.135	4	9	0.0000	2.34
Polluting firms	POLLUTING	0.274	0.0000	1	0.0000	0.446
Financing costs	FC	0.105	0.076	0.981	0.0000	0.108
Operational efficiency	TURN	0.938	0.804	5.144	0.014	0.635
Accounting information quality	ABSDA	0.106	0.075	0.936	0.0002	0.108
Firm age	AGE	1.561	1.602	1.857	1.146	0.175
Board size	BOARD	5.011	5	7	5	0.149
Environmental responsibility disclosure	DISCER	0.336	0.0000	1	0.0000	0.472
Financial leverage	LEV	0.596	0.59	2.956	0.046	0.285
Fixed assets	PPE	0.243	0.209	0.932	0.003	0.162
Firm size	SIZE	14.391	14.292	20.14	10.532	1.453
State of ownership	STAT	0.656	1	1	0.0000	0.475

gives the descriptive statistics of the variables.

It must be mentioned that panel data with a fixed-effects test were used (based on Chow’s and Hausman’s tests, which revealed a significance below 5% for both research models).

In the current research, six equations were used to test the hypotheses. Chow’s test was used to test the selection between the pooled and panel data types. Also, to determine whether the method used in the model estimation was a fixed or random effects, Hausman’s test was used. The results of these tests are shown in panel A and panel B in

Table 3. The results of Chow’s and Hausman’s tests.

Panel A: The Results of Chow’s Test
Equation number	F Stat.	p-value	Results
Equation (4)	36.142	0.000	Panel data
Equation (6)	6.963	0.000	Panel data
Equation (8)	8.836	0.000	Panel data
Panel B: The results of Hausman’s test
Equation number	$X^2$ Stat.	p-value	Results
Equation (4)	375.03	0.000	Fixed effects
Equation (6)	366.087	0.000	Fixed effects
Equation (8)	322.923	0.000	Fixed effects

.

It should be noted that the results show that all the models are panel data with fixed effects. It should be pointed out that the dynamic panel method has been used for other models, considering that the research data are in the form of panel data. The generalized method of moments (GMM) approach was used to solve the endogeneity problem. In this regard, two serial correlation tests and an instrument validity test (Hansen test) were examined beforehand, and the results show that there is no serial correlation between the errors. Also, the results of the Hansen test (which was more than 5%) show that the validity of the tools is confirmed.

4.2. Results of Testing the Basic Research Model

Table 4. Testing the basic research equation.

Variable	Symbol	Est. Coeff.	t Stat.	Sig.
Constant	C	0.629	1.849	0.0649
Return on assets	ROAt	0.156	4.373	0.0000
-	POLLUTING*ROAt	−0.486	−6.848	0.0000
-	CSR*POLLUTING*ROAt	0.045	5.395	0.0000
Accounting information quality	ABSDA	0.109	3.483	0.0005
Firm age	AGE	−0.32	−2.566	0.0105
Board size	BOARD	−0.009	−0.836	0.4032
Environmental responsibility disclosure	DISCER	−0.046	−1.969	0.0493
Financial leverage	LEV	−0.426	−24.11	0.0000
Fixed assets	PPE	−0.356	−9.477	0.0000
Firm size	SIZE	0.033	4.226	0.0000
State of ownership	STAT	−0.161	−5.318	0.0000
R2	0.896	F stat.	45.54
R2 adjusted	0.877	Sig.	0.0000
Durbin–Watson		1.532

indicates the results of testing Equation (1) (the basic equation used in this study).

As Table 4 shows, the significance level of the corporate social responsibility variable is 0.000 for the firms in polluting industries, which is smaller than the significance level considered in the present work (5%), and indicates the significant and positive influence of this variable on the persistence of earnings at the 95% confidence level. Thus, fulfilling environmental responsibilities leaves a significant and positive effect on the persistence of earnings of firms in polluting industries, as the coefficient of ${\mathsf{\beta }}_2$ is larger than zero. The first and second hypotheses will be tested based on the following results.

4.3. Results of the Testing of Hypothesis I

To test the first hypothesis, the results of Equation (2) need to be examined, considering the moderating role of the asset turnover variable (as an index of operational efficiency). Hypothesis I will be confirmed provided that the coefficient of fulfilling the CSR of the polluting firms (${\mathrm{m}}_2)$ is significant and positive in the first equation, so that the CSR of the polluting firms leads to increased operational efficiency, and the coefficient of operational efficiency (${\mathsf{\beta }}_1$) is significant and positive in Equation (3). In such a case, fulfilling the CSR will enhance the earnings persistence of polluting firms through increased operational efficiency. Equations (2) and (3) are based on the operational efficiency variable. We will call them Equations (4) and (5).

$${\mathrm{TURN}}_{\mathrm{it}}=\complement +{\mathrm{m}}_1{\mathrm{POLLUTING}}_{\mathrm{it}}+{\mathrm{m}}_2{\mathrm{CSR}}_{\mathrm{it}}\times {\mathrm{POLLUTING}}_{\mathrm{it}}+{\mathrm{m}}_3{\mathrm{CSR}}_{\mathrm{it}}+\mathsf{\beta }{\mathrm{X}}_{\mathrm{it}}+\mathsf{\epsilon }$$

(4)

$${\mathrm{ROA}}_{\mathrm{it}+1}=\complement +\mathsf{\rho }{\mathrm{ROA}}_{\mathrm{it}}+{\mathsf{\beta }}_1{\mathrm{TURN}}_{\mathrm{it}}\times {\mathrm{ROA}}_{\mathrm{it}}+{\mathsf{\alpha }}_2{\mathrm{CSR}}_{\mathrm{it}}\times {\mathrm{POLLUTING}}_{\mathrm{it}}+\mathsf{\beta }{\mathrm{X}}_{\mathrm{it}}+\mathsf{\epsilon }$$

(5)

As

Table 5. Results of testing model 4 (based on the operational efficiency variable).

${\mathbf{TURN}}_{\mathbf{it}}=\complement +{\mathbf{m}}_1{\mathbf{POLLUTING}}_{\mathbf{it}}+{\mathbf{m}}_2{\mathbf{CSR}}_{\mathbf{it}}\times {\mathbf{POLLUTING}}_{\mathbf{it}}+{\mathbf{m}}_3{\mathbf{CSR}}_{\mathbf{it}}+\mathsf{\beta }{\mathbf{X}}_{\mathbf{it}}+\mathsf{\epsilon }$
Variable	Symbol	Est. Coeff.	t Stat.	Sig.
Constant	C	3.11	3.23	0.0013
Polluting industries	POLLUTING	0.39	3.11	0.0019
Corporate social responsibility of polluting industries	CSR*POLLUTING	0.008	2.001	0.0484
Corporate social responsibility	CSR	−0.02	−0.78	0.4335
Accounting information quality	ABSDA	0.06	2.09	0.0366
Firm age	AGE	0.61	1.39	0.1621
Board size	BOARD	−0.01	−0.64	0.5202
Environmental responsibility disclosure	DISCER	0.01	0.22	0.8187
Financial leverage	LEV	−0.16	−3.53	0.0004
Fixed assets	PPE	−0.82	−9.59	0.0000
Firm size	SIZE	−0.17	−6.36	0.0000
State of ownership	STAT	−0.32	−12.27	0.0000
R2	0.87	F stat.	37.99
R2 adjusted	0.85	Sig.	0.0000
Durbin–Watson		1.516

shows, the significance level of the CSR variable is 0.04 for the firms in polluting industries, which is smaller than the significance level considered in the present work (5%), and indicates the significant and positive influence of this variable on the asset turnover index at the 95% confidence level. Thus, fulfilling their corporate social responsibilities leaves a significant and positive effect on the asset turnover of firms in polluting industries, as the coefficient of the corporate social responsibility of polluting firms is positive (0.008). On the other hand,

Table 6. Results of testing model 5 (based on the operational efficiency variable).

${\mathbf{ROA}}_{\mathbf{it}+1}=\complement +\mathsf{\rho }{\mathbf{ROA}}_{\mathbf{it}}+{\mathsf{\beta }}_1{\mathbf{TURN}}_{\mathbf{it}}\times {\mathbf{ROA}}_{\mathbf{it}}+{\mathsf{\alpha }}_2{\mathbf{CSR}}_{\mathbf{it}}\times {\mathbf{POLLUTING}}_{\mathbf{it}}+\mathsf{\beta }{\mathbf{X}}_{\mathbf{it}}+\mathsf{\epsilon }$
Variable	Symbol	Est. Coeff.	t Stat.	Sig.
Constant	C	0.6005	1.887	0.0595
Return on assets	ROAt	0.002	0.033	0.9736
-	POLLUTING*ROAt	0.108	2.23	0.0261
-	CSR*POLLUTING*ROAt	0.017	1.625	0.1047
Accounting information quality	ABSDA	0.108	3.749	0.0002
Firm age	AGE	−0.327	−2.584	0.01
Board size	BOARD	−0.008	−0.782	0.434
Environmental responsibility disclosure	DISCER	−0.051	−2.181	0.0295
Financial leverage	LEV	−0.406	−19.993	0.0000
Fixed assets	PPE	−0.33	−8.104	0.0000
Firm size	SIZE	0.032	4.098	0.0000
State of ownership	STAT	−0.158	−5.549	0.0000
R2	0.896	F stat.	45.44
R2 adjusted	0.877	Sig.	0.0000
Durbin–Watson		1.5077

indicates the significance level of assets turnover to be 0.026, which is smaller than the significance level considered in the present work (5%), and indicates the significant and positive influence of this variable on the persistence of earnings at the 95% confidence level. Regarding the direction of influence, the coefficient of this variable is 0.108, suggesting a positive effect. Hence, the first hypothesis, which suggests a relationship between the fulfillment of CSR by firms and the persistence of earnings with the moderating role of operational efficiency is confirmed, as ${\mathsf{\beta }}_1$ and ${\mathrm{m}}_2$ are both positive and significant.

4.4. Results of the Testing of Hypothesis II

Given the moderating role of financing costs, the results of Equations (3) and (8) need to be examined to test the second hypothesis. The second hypothesis, which suggests that the fulfillment of corporate social responsibility improves the persistence of earnings by reducing the financing costs of polluting firms will be confirmed if the coefficient of CSR in the polluting firms, ${\mathrm{m}}_2$, is significant and negative in Equation (3) (suggesting that fulfilling their corporate social responsibility reduces the financing costs of polluting firms), and if the coefficient of financing costs, ${\mathsf{\beta }}_1$, is significant and positive. Equations (2) and (3) are based on the financing costs variable.

$${\mathrm{FC}}_{\mathrm{it}}=\complement +{\mathrm{m}}_1{\mathrm{POLLUTING}}_{\mathrm{it}}+{\mathrm{m}}_2{\mathrm{CSR}}_{\mathrm{it}}\times {\mathrm{POLLUTING}}_{\mathrm{it}}+{\mathrm{m}}_3{\mathrm{CSR}}_{\mathrm{it}}+\mathsf{\beta }{\mathrm{X}}_{\mathrm{it}}+\mathsf{\epsilon }$$

(6)

$${\mathrm{ROA}}_{\mathrm{it}+1}=\complement +\mathsf{\rho }{\mathrm{ROA}}_{\mathrm{it}}+{\mathsf{\beta }}_1{\mathrm{FC}}_{\mathrm{it}}\times {\mathrm{ROA}}_{\mathrm{it}}+{\mathsf{\alpha }}_2{\mathrm{CSR}}_{\mathrm{it}}\times {\mathrm{POLLUTING}}_{\mathrm{it}}+\mathsf{\beta }{\mathrm{X}}_{\mathrm{it}}+\mathsf{\epsilon }$$

(7)

As

Table 7. Testing the model 6 (based on the financing costs variable).

${\mathbf{FC}}_{\mathbf{it}}=\complement +{\mathbf{m}}_1{\mathbf{POLLUTING}}_{\mathbf{it}}+{\mathbf{m}}_2{\mathbf{CSR}}_{\mathbf{it}}\times {\mathbf{POLLUTING}}_{\mathbf{it}}+{\mathbf{m}}_3{\mathbf{CSR}}_{\mathbf{it}}+\mathsf{\beta }{\mathbf{X}}_{\mathbf{it}}+\mathsf{\epsilon }$
Variable	Symbol	Est. Coeff.	t Stat.	Sig.
Constant	C	1.479	9.996	0.0000
Corporate social responsibility polluting firms	POLLUTING	−0.013	−0.879	0.3797
-	CSR*POLLUTING	0.011	1.393	0.1641
Corporate social responsibility	CSR	−0.016	−4.252	0.0000
Accounting information quality	ABSDA	−0.009	−0.591	0.5541
Firm age	AGE	−0.823	−11.547	0.0000
Board size	BOARD	−0.008	−0.68	0.4966
Environmental responsibility disclosure	DISCER	0.067	1.806	0.0713
Financial leverage	LEV	0.038	3.355	0.0008
Fixed assets	PPE	0.026	1.136	0.2561
Firm size	SIZE	−0.001	−0.276	0.7822
State of ownership	STAT	−0.027	−0.318	0.7503
R2	0.639	F stat.	9.523
R2 adjusted	0.572	Sig.	0.0000
Durbin–Watson		1.502

shows, the significance level of the CSR variable is 0.164 for the firms in polluting industries, which is larger than the significance level considered in the present work (5%), and indicates the insignificant and positive influence of this variable on financing costs at the 95% confidence level. On the other hand,

Table 8. Testing the model 7 (based on the financing costs variable).

${\mathbf{ROA}}_{\mathbf{it}+1}=\complement +\mathsf{\rho }{\mathbf{ROA}}_{\mathbf{it}}+{\mathsf{\beta }}_1\mathbf{FC}\times {\mathbf{ROA}}_{\mathbf{it}}+{\mathsf{\alpha }}_2{\mathbf{CSR}}_{\mathbf{it}}\times {\mathbf{POLLUTING}}_{\mathbf{it}}+\mathsf{\beta }{\mathbf{X}}_{\mathbf{it}}+\mathsf{\epsilon }$
Variable	Symbol	Est. Coeff.	t Stat.	Sig.
Constant	C	0.577	1.803	0.0715
Return on assets	ROAt	0.086	2.042	0.0415
-	FC*ROAt	0.248	2.279	0.023
-	CSR*POLLUTING	0.015	1.472	0.1415
Accounting information quality	ABSDA	0.104	3.671	0.0003
Firm age	AGE	−0.299	−2.204	0.0279
Board size	BOARD	−0.009	−0.896	0.3702
Environmental responsibility disclosure	DISCER	−0.049	−2.186	0.0291
Financial leverage	LEV	−0.42	−24.604	0.0000
Fixed assets	PPE	−0.34	−8.852	0.0000
Firm size	SIZE	0.0319	4.142	0.0000
State of ownership	STAT	−0.161	−5.854	0.0000
R2	0.893	F stat.	45.069
R2 adjusted	0.873	Sig.	0.0000
Durbin–Watson		1.511

indicates the significance level of financing costs to be 0.023, which is smaller than the significance level considered in the present work (5%), and indicates the significant and positive influence of this variable on the persistence of earnings at the 95% confidence level. The second hypothesis is thus rejected, as ${\mathrm{m}}_2$ is insignificant.

5. Additional Tests

Managers have high levels of accounting discretion regarding corporate social responsibility (CSR) expenditure under the current financial reporting framework because of the lack of accounting standards which address CSR expenditure directly [79]. The ERC measures the relation between unexpected stock returns and unexpected earnings (UEs) [80]. The ERC represents the consistency of information reflected in the two variables—stock returns and earnings. Thus, the CSR–ERC relation is expected to help evaluate how well accounting earnings reflect CSR implications. Ohlson (1995) believed if the CSR could enhance companies’ earnings persistence, it would also improve their investors’ earnings response [77].

A third hypothesis suggesting a relationship between the fulfillment of corporate social responsibility and the earnings response coefficients (ERC) was examined with further tests. Feltham and Ohlson’s residual earnings model [81] was used to calculate the ERC.

$${\mathrm{MV}}_{\mathrm{it}+1}=\complement +\mathsf{\rho }{\mathrm{ROA}}_{\mathrm{it}}+{\mathsf{\beta }}_1{\mathrm{POLLUTING}}_{\mathrm{it}}\times {\mathrm{ROA}}_{\mathrm{it}}+{\mathsf{\beta }}_2{\mathrm{CSR}}_{\mathrm{it}}\times {\mathrm{POLLUTING}}_{\mathrm{it}}\times {\mathrm{ROA}}_{\mathrm{it}}+\mathsf{\beta }{\mathrm{X}}_{\mathrm{it}}+\mathsf{\epsilon }$$

(8)

MV is the response of the stakeholders to the earnings, which is the ratio of the book to market value.

As

Table 9. Testing the third research hypothesis.

Variable	Symbol	Est. Coeff.	t Stat.	Sig.
Constant	C	15.38	4.288	0.0000
Return on assets	ROAt	−0.008	−0.064	0.9484
-	POLLUTING*ROAt	2.053	3.038	0.0025
-	CSR*POLLUTING*ROAt	−0.29	−2.017	0.0441
Accounting information quality	ABSDA	−0.286	−2.228	0.0262
Firm age	AGE	−10.077	−4.215	0.0000
Board size	BOARD	0.011	0.09	0.9276
Environmental responsibility disclosure	DISCER	0.224	5.716	0.0000
Financial leverage	LEV	0.276	4.058	0.0001
Fixed assets	PPE	0.34	6.723	0.0000
Firm size	SIZE	0.062	1.274	0.2031
State of ownership	STAT	0.068	0.541	0.5886
R2	0.68	F stat.	11.454
R2 adjusted	0.621	Sig.	0.0000
Durbin–Watson		1.50

and the model for the third research hypothesis indicate, the significance level of corporate social responsibility fulfillment in the polluting firms on the ERC was 0.0441, which is smaller than the significance level considered in the present work (5%), and indicates the insignificant and positive influence of this variable on the ERC at the 95% confidence level. Regarding the direction of effect, the impact is negative as the coefficient is equal to −0.29, suggesting that the increased fulfillment of corporate social responsibility by polluting firms reduces the ERC. It can thus be inferred that there is a significant relationship between CSR fulfillment in firms and the ERC, confirming the third research hypothesis.

6. Sensitive Analysis

The effects of COVID-19 have been severe around the world and are still unfolding. In addition to the increased number of infections and deaths associated with this epidemic and increased household poverty [82], the economic consequences have included reduced GDP, increased unemployment, and the loss of thousands of businesses [83].

These economic consequences have made stock markets and oil prices more volatile [84,85] and have affected stock returns [86] and investors’ risk perceptions [87]. The COVID-19 pandemic has directly affected incomes due to absenteeism and reduced productivity, creating a negative supply shock and slowing manufacturing activity due to global supply chain disruptions and factory shutdowns. For example, in China, the manufacturing index in February 2020 fell by more than 54% from the previous month’s value [88].

Therefore, the present study sought to investigate the effect of an epidemic such as COVID-19 on the efficiency and performance of companies. Hence, the hypotheses examined for 2014–2019 were tested again, the results of which are shown in

Table 10. Retesting the basic research equation.

Variable	Symbol	Est. Coeff.	t Stat.	Sig.
Constant	C	1.534	6.399	0.0000
Return on assets	ROAt	0.076	2.018	0.0441
-	POLLUTING*ROAt	−0.550	−7.330	0.0000
-	CSR*POLLUTING*ROAt	0.052	3.896	0.0000
Accounting information quality	ABSDA	0.060	4.430	0.0000
Firm age	AGE	−0.643	−4.671	0.0000
Board size	BOARD	−0.014	−2.429	0.0155
Environmental responsibility disclosure	DISCER	−0.069	−4.450	0.0000
Financial leverage	LEV	−0.462	−35.899	0.0000
Fixed assets	PPE	−0.342	−8.757	0.0000
Firm size	SIZE	0.008	1.365	0.1727
State of ownership	STAT	−0.164	−6.189	0.0000
R2	0.926	F stat.	56.25
R2 adjusted	0.910	Sig.	0.0000
Durbin–Watson		1.597

.

As Table 10 shows, the significance level of the corporate social responsibility variable is 0.000 for the firms in polluting industries, which is smaller than the significance level considered in the present work (5%), and indicates the significant and positive influence of this variable on the EP at the 95% confidence level. Thus, fulfilling their environmental responsibilities leaves a significant and positive effect on the EP of firms in polluting industries as the coefficient of ${\mathsf{\beta }}_2$ is larger than zero.

It should be noted that the test results of the hypotheses and models have shown that the results would not change significantly. In other words, the results obtained before and after COVID-19 are the same. Also, due to the length of this study, we decided to not present additional tables.

7. Research Discussion and Conclusions

Many academic studies have examined the relationship between corporate social responsibility (CSR) and earnings quality. Firms that engage in socially responsible activities tend to have a higher earnings quality as they tend to have adopted better corporate governance [89], have higher disclosure and transparency levels, and have a lower risk of suffering from a damaged reputation [90]. Moreover, the firms engaged in CSR may have better access to financing and lower capital costs due to their better reputations and the trust of their stakeholders. Still, some studies have reported contrasting or uncertain results on the relationship between earnings quality and CSR. This might be due to the difference in CSR and earnings management measurement methods and the different distinctions made among the studied countries and industries [91,92]. Evidence suggests that CSR could positively affect earnings management (ERC and persistence) [45,80,93]; however, this relationship is complex and may be influenced by variable factors. Further research appears to be needed to understand better the mechanisms and conditions under which CSR enhances earnings persistence [94].

The testing of the first research hypothesis revealed a significant relationship between the fulfillment of corporate social responsibility by firms and the persistence of earnings, with the moderating role of operational efficiency [95,96] This means that the fulfilling of corporate social responsibility by firms in polluting industries increases operational efficiency, ultimately leading to earnings persistence. The results of the testing of the first research hypothesis can be expressed as operational efficiency improvement with the fulfillment of CSR. It requires enhancing production technology, higher staff productivity, and more efficient supply chains. Furthermore, higher operational efficiency is a competitive advantage that eventually affects a firm’s performance and earnings persistence [97]. The results of the testing of the first research hypothesis agree with the results of [2,98].

The results of the testing of the second research hypothesis revealed no significant relationship between the fulfillment of corporate social responsibility by firms and the persistence of earnings, with the moderating role of financing costs. The lack of a significant relationship between the corporate social responsibility fulfillment by firms and financing costs could have many reasons, such as the inability of Iranian firms to turn a better social performance into higher profitability. The lack of a relationship between these two variables over the short term was rather expected, as social responsibility’s effects are inherently long-term. The results of the testing of the second variable are inconsistent with those of [2].

The results of the third research hypothesis revealed no significant relationship between the fulfillment of CSR by firms and the ERC. Social responsibility disclosure in annual reports is one of the ways for a firm to show its shareholders that it can manage them well and enhance business performance. Management reasons that there is a high correlation between good management practice and socially responsible activities, since participation in such activities improves the relationship between the firm and its shareholders. Fulfilling its social responsibility enables firms to create common ground between their own and their shareholders’ interests [99]. Hence, social responsibility directs financial support toward financial enterprises, which they can use to fight operational uncertainty, reduce the income instability resulting from credit market changes, improve their earnings persistence, and positively affect the ERC. The results of the testing of the second hypothesis agree with the results of [2,98].

Environmental accounting provides the necessary (monetary) information for managers to make decisions regarding the business unit’s control process and environmental performance improvement [100]. The strategy studied in the present work concerns future threats and changes that could impose new and greater costs on this enterprise. The adopted operational strategy deals with costs imposed on a firm, the breadth and amount of which the firm is unaware. Through this strategy, the firm can save costs and manage expenses. Environmental costs resulting from pressures from agents outside the company, such as the government and customers, are expected to be imposed on firms, as shown by [101]. Hence, environmental accounting can help improve the environmental performance of firms, by reducing the pressure from the government and customers on the one hand, and by contributing to the financial and operational indices through optimal resource allocation on the other [102,103].

8. Limitations and Suggestions for Future Research

Our study has produced rigorous and robust results, but we also acknowledge its limitations. Our sample size was limited as we could only collect CSR and financial data from the most reputable companies in Iran. To improve the generalizability of our findings, we encourage future research to include observations from all Iranian companies. This study’s primary limitation is related to the CSR measure. Measures of CSR use different methods and depend on the geographical area, culture, and other factors. Therefore, the importance of some environmental disclosures may differ from country to country.

It would be advisable to conduct additional research that could expand the scope of this study beyond a single country to a cross-national setting, such as the Middle East and North Africa (MENA) region. Furthermore, although our CSR indices are based on their quantity and quality, the analysis could be enhanced by using other CSR indicators in future studies.

Another limitation of this research is the confounding factors that were not seen in the present study due to the unavailability of data, such as the increasing importance of environmental, social, and governance (ESG) activity and monetary expansion. Investors may pay a premium for socially responsible investments, and should invest with a long-term perspective, according to the theoretical work on ESG [104]. Empirical studies have suggested that socially responsible funds exhibit less sensitivity to past negative returns, which aligns with the hypothesis of longer investment horizons [105,106]. Regarding the other confounding factor, namely monetary expansion, the COVID-19 crisis caused, starting in March 2020, the Fed and 20 other central banks to put forth unconventional monetary policy measures to respond to the severe recession caused by the COVID-19 pandemic [107]. According to Rebucci et al. [108], the effectiveness of quantitative easing (QE) has not decreased in advanced economies.

Furthermore, the international transmission of QE is consistent with the operation of a long-run uncovered interest rate parity. During the COVID-19 pandemic, a significant dollar shortage shock was observed. In emerging economies, the impact of QE on bond yields is much more substantial, and its transmission to exchange rates differs from that in advanced economies. As an emerging economic environment, Iran could not implement proper expansionary economic policies due to economic sanctions and the outbreak of COVID-19. The intensification of instabilities in 2020, i.e., the beginning of COVID-19, the sharp decrease in oil exports (due to the tightening of sanctions), the decrease in non-oil exports (due to the spread of the coronavirus), and the increase in exchange rates, triggered inflationary expectations [108]. The increase in inflation expectations caused liquidity to flow into the assets market (real estate, stocks, durable goods, etc.) and became the basis for inflammation in these markets. In this situation, the central bank, with a delay in its monetary policy, did not use the available tools to reduce the macroeconomic fluctuations, and the sharp decrease in the interbank interest rate intensified the fluctuations. Therefore, for Iran’s environment, it is impossible to analyze the above-mentioned hindering factors, so it is suggested that researchers study these aspects in future works.

Based on the results of this study, it is suggested that managers develop proper social responsibility programs to provide a suitable framework with which to integrate social responsibility programs and activities with accounting information quality. Moreover, it is recommended that the Securities and Exchange Organization use conventional models to develop standards to rate social responsibility disclosure levels for the firms operating in the capital market, given the information content of social responsibility for this market.