Are Socially Responsible Companies Really Ethical? The Moderating Role of State-Owned Enterprises: Evidence from China

Abstract

The present research aims to explore the relationship between corporate social responsibility and earnings management (EM). For this study, we utilized the panel data of companies registered with the Shanghai and the Shenzhen stock exchanges. The data consists of 10years of financial data from 2010 to 2019. After a thorough investigation, we discovered that CSR hurts the EM practice, which mainly relates to the prevalence of the ethical stance and the moral stance in corporate decision-making. When firms engage in activities about CSR, they tend to improve their corporate image and their social image as the stakeholder satisfaction level increases. The results also indicate that, when firms engage in these types of activities, they tend to incorporate practices related to CSR as part of their corporate strategy. This also results in a higher moral standing amongst the decision-makers, and they prefer to reject malpractices, such as EM, as a result. In the case of the Chinese state-owned firms, the results indicate that these companies increasingly engage in real earnings management (REM), even though they have increased their CSR activities. The results point towards management opportunism with Chinese state-owned companies.

1. Introduction

The financial crises of 2007–2008, which was coupled with financial scandals, has created a great deal of mistrust amongst the different stakeholders in the economies around the world. This has caused a great deal of distress for companies and in order to regain the trust of the stakeholders, the companies have started to commit themselves towards social responsibility and have started to consider CSR, ethical behaviour, and disclosures relating to CSR as a vital part of their reporting mechanisms [1]. They realized that they not only needed to survive, but they also need to ensure growth and the future stability of their revenues along with their financial positions. Ameer [2] states that, by keeping these factors in mind, the companies have considered comprehensive environmental alertness, ethical behavior, and practices. The past research is inconclusive regarding whether CSR and ethical practices have a positive or a detrimental impact on a business’s reporting quality [3]. Shleifer [4] contended that socially responsible companies are less likely to undertake earnings management (EM) activities, because they can have a determinant impact on their established reputation.

In recent times, the corporations have changed their perspective regarding financial assistance in the form of corporate donations and are moving towards the betterment of the overall community by involving their employees. According to Porter and Kramer [5], corporations are coming up with charitable programs that are designed in such a way that it not only involves the community but also satisfies the employees; as a result, not only does the reputation of the business improve, but the morale of employees also improves, along with improvements in the environment and overall living standard of the community. Along this line, Lewin and Sabater [6] state the example of IBM, who has launched programs such as social volunteering, which has led to the improvement of not only the corporate reputation but also overall employee satisfaction.

There is consensus amongst researchers and academics that such programs are only successful if they are in line with the strategic vision and objectives of the companies. Along with that, the social activities selected for such programs and their potential impact on society or the environment needs to be communicated to such volunteers. This not only motivates the employees involved in the activity but also improves their commitment towards the cause and the organization as a whole. According to Brewis [7], such exercises not only improve employee satisfaction but also result in positive influences that reaffirm employee loyalty, as it aligns with that of the organization. He further states that this also has a positive impact on the overall corporate image of the organization in the eyes of both employees and society as a whole.

Castelo Branco and Lima Rodriques [8] state that CSR and ethics are related to the characteristics of corporate decision-making that are related to moral issues, such as environmental protection, practices related to the wellbeing of human resources, and helping deprived communities establish good working relations with customers and employees. CSR engagements by companies improve their social profiles and enhance their corporate reputations [9], which in turn not only satisfies the stakeholders but also reduces the financial risks for companies involved in these types of activities [9].

Over the last 30 years, the Chinese economy has experienced phenomenal growth. China has attained the status of the second largest economy in the world by surpassing Japan in 2017, and the Chinese economy is expected to surpass the USA in the next decade. According to Keely and Anderson [10], a report from the international monetary fund (IMF) showed that the Chinese economy cannot sustain its current structure as an export-oriented economy for much longer, and it needs to convert itself into a consumer-oriented economy if it wants to retain its economic status (IMF2017). It is transforming itself, and it is fast becoming a consumer-oriented economy. The Chinese firms are becoming multinational firms and foreign investment is pouring intothe companies that have acknowledged the importance of CSR and ethical practices. One has to keep in mind that the need for CSR and ethical activities is mainly driven by the domestic stakeholders, and their practices are more oriented towards Chinese domestic audiences. This is an important factor in analyzing the CSR and ethical activities conducted by Chinese companies. They are considered as a part of a larger national development plan [11,12].

These activities bare telltale signs of the government influence of CSR and ethical activities. The reason could be attributed to past issues faced by the Chinese exporters who were penalized for a lack of CSR activities by developed nations [12,13]. With changes in the legal setup, it has now become imperative for all companies to perform. This intermediate transformation in the Chinese CSR and ethical practices becomes more substantial, and other emerging markets are emulating the Chinese growth model CSR practices [14,15].

Even though there has been plenty of research on CSR in China, it can be considered irrelevant for two main reasons. A major portion of the research has utilized the accrual earnings management (AEM) (discretionary accruals) approach to measure the EM [3,16,17,18,19]. This can easily be manipulated by changing the operating activities rather than manipulating the discretionary accruals. This technique is commonly known as real earnings management (REM) [20,21]. Kim et al. [21] stated that the REM-based approach has seldom been applied in the Chinese context, and a similar view is held by Bozzolan et al. (2015) [22]. Nonetheless, Kuo and Ning [23] imply that Kim et al. [21] and Bozzolan et al. [22] held that the conclusions based upon AEM lack consistency with the Chinese EM practices. The second reason relates to the institutional factors, which are essentially varied from other countries when it comes to CSR and ethical practices in China. The sample size of the previous research was very limited, and it accounted for no more than five percent of the entire firms population [3]. These researchers are not alone, because Scholtens and Kang (2013) [17] and Bozzolan et al. (2015) [22] also support similar notions.

This study mainly focuses on the relationships between CSR and EM practices in the context of companies based and operating in China. Based on our research study, we concluded that socially responsible companies in China are less likely to engage in AEM and REM activities. The major contributions of this study are as follows.

Our interest in China was motivated for socio-economic reasons. Firstly, China is the biggest developing nation and the world’s second largest economy after the USA. China also has problems associated with developing nations [24,25]. Of particular relevance to this point, it was rightly observed [26,27,28,29,30] that few studies on CSR and EM have been conducted outside developed nations such as USA and UK, which very clearly enjoyed corporate governance, legal, economic, and institutional contexts. However, developing economies, such as the Chinese, have received little attention. It is strongly argued that their different corporate governance CG, legal, and economic environments deserve to be investigated. The degree to which formal mechanisms of CG can contain a managerial ability to engage in EM may differ. Moreover, the importance of emerging markets has been addressed in several studies [24,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52], and China is the largest economy offering an interesting context.

Secondly, as reported by Li et al. [53], the relation between CSR and REM has not yet been thoroughly investigated. For instance, there is a relative dearth of research on the impact of CSR on EM. Achleitner et al. [54], Zang [55], and Cohen and Zarowin [56] agree that most previous investigations of EM have focused on manipulating discretionary accruals and might be very productive to investigate EM as a technique to manipulate real earnings. Managers have been frequently observed using both practices alternatively. Cohen and Zarowin [56], Francis et al. [57] argued that, after the 2002 Sarbanes–Oxley Act (SOX), real EM became the central attraction of managers and researchers instead of accrual EM. These scholars explained this shift of interest by the fact that real EM is costly and harder to deduct from regulators and auditors.

Third, many previous studies on this topic have been inconclusive, or they presented contradictions with the relationship between CSR and EM. These studies mainly include [3], and their findings were further endorsed in [16]. Meanwhile, Kim et al. [21], which was followed by [58], also held similar conclusions, and these were supported by the research in [22]. Even though studies by Noronha et al. [59] had provided circumstantial evidence aimed at the relationship between CSR and EM in the Chinese context, which was followed by [15], there were not any conclusive results that were provided. The study [14] also suffers from similar issues. Meanwhile, our study provides conclusive evidence regarding a negative relationship between CSR and EM.

Fourth, rather than relying on one method to measure EM, we applied both the AEM and the REM methods. By following both methods, we not only validated our findings, but we were able to check the impact of CSR on AEM and REM distinctly. It is generally held that the political costs and the acceptability of issues relating to AEM are higher than those of REM. In [55], it was explained that the main reason for this relates to the fact that AEM is the focus of an examination by auditors during annual audits and regulators alike [60].

Fifth, companies in China practice a wide variety of CSR activities, and they have experienced a great degree of changes over time at the corporate and the institutional levels. The rise of the Chinese economy along with its stable growth pattern has provided ideal circumstances to observe the external institutional mechanisms discussed in [12]. Yin [61] also researched the same topic, which was followed by Islam [15] and Parsa et al. [14], and they all have similar conclusions. These researchers have also pointed out that the CSR and ethical behaviour practices of Chinese companies conform to their social and cultural practices of creating harmony in society. This might seem to differ from the conventional practices and objectives of Western companies, but they conform to wider Asian and South East Asian societies. As a result, this research can be considered as a stepping stone for further research into the topic, especially in these countries.

This paper follows the following pattern. In the second section, a detailed literature review is presented. We formulated our hypothesis based upon the literature review. The third section includes a research design, and the methods that are implemented and the choice variables are discussed in detail. The fourth section provides the results. The last section contains the discussion and the implications of the research, followed by the conclusion.

2. Theoretical Framework, Literature Review, and Research Hypothesis

2.1. Corporate Social Responsibility and Earning Management

Over the last three decades, the relationship between CSR and the profitability of companies has received substantial attention [62,63,64,65]. The bases for CSR can be traced to the stakeholder theory [66]. These academics have stressed that commercial enterprises need to have good relationships with all of the stakeholders [67]. Many scholars [62,68] in line with these viewpoints have argued that this not only enhances the reputation but also has a positive impact on the financial performance. Pesqueux and Damak-Ayadi [69] had stated that CSR activities are also indicative of their future profitability and growth.

We based our study on the stakeholder theory. This theory explains why the companies are willing to spend their precious resources on activities that do not seem to have any returns in the short term. These activities are continuously enhancing the reputation of the companies, which usually results in a positive image, and the management can utilize this image to increase the revenues in the long term. This results in the financial stability of the companies, which is a key success factor for any business [9]. Along with these activities, the companies can incorporate the concerns and knowledge of these stakeholders into their long-term strategies, which provides stable platforms for companies to expand and increase their market share.

To achieve the basic and long-term objectives, the companies need to build and sustain positive relationships with all the stakeholders, because they control resources, which are vital for the growth and the survival of businesses. Choi and Lee [58] state that the companies that take a greater part in CSR activities will have a positive image built over time and at the cost of precious resources, and they are more likely to be transparent in their financial reporting.

According to [70], the CSR is positively correlated to fair financial disclosure, which is further endorsed in [71]. Meanwhile, in [72,73], the authors contended that the CSR activities can reduce the agency problems that exist amongst the different stakeholders within and outside the company. Jones [67] cites the institutional stakeholder theory, and his recommendations are also in line with those previously mentioned. The fundamental reason cited in [3] is that the perception of a socially responsible company is favorable in the eyes of the general public, and they tend to think that these types of companies are not hiding any facts from them.

Because EM is considered to be an anti-social activity, it deceives the users of financial statements. It is generally believed that socially responsible firms have more involvement from the stakeholders, and it becomes hard for them to conduct such activities. The incentive for management in these types of companies to indulge in these types of activities is very low. This is indicative of a negative relationship between CSR and EM.

According to Gelb and Strawser [74], these firms are always looking for long-term objectives, and they prefer to foster a potentially profitable relationship with the stakeholders that is in line with the long-term perspective hypothesis. These views are also backed by the researchin [58]. Hong and Andersen [75] argued that the commitment of companies to CSR reduces their inclination towards malpractices, such as EM. As per the findings of [3,74], we assumed that socially responsible firms are less likely to be involved in EM and vice versa. Based on the above discussion, we propose our first hypothesis.

H1. 

The extent of CSR is negatively associated with the level of EM.

2.2. The Moderating Role of State Control Enterprises on CSR and EM

The studies in the past have focused on the relationship between CSR and the financial performance of the firms [9,76], but the results reported have been mixed. Jorgensen and Knudsen (2006) further noted that this association signifies the most interrogated area of CSR [77]. Many researchers have found evidence in favor of a positive relationship, but the strength of the relationship is rather weak according to several scholars [64,78,79]. Several researchers [80,81] have also stated that researching this connection cannot be deemed definite. According to Jawahar and McLaughlin [82], the underlying mechanisms, which impact the financial performances, need to be understood in the local context [83]. To summarize the literature review, we can safely say that there seems to be a lack of understanding regarding the impact of CSR initiatives, which have an impact on the quality of reported earnings and on the overall financial performance of a company. In addition, the role of state ownership in firms in emerging markets has been extensively discussed in several papers [43,47,51,84]. Our present study aims to bridge this gap by studying the moderating role of state-owned enterprises on the CSR and EM relationship.

Keeping in mind that we paid special consideration to the inherent limitations of CSR practices adopted by the Chinese firms. In our opinion, these can have a moderating impact with the relationship between CSR and EM. One of the main factors in this regard is state control. Hung [85] argued that the government has drafted tough laws and regulations to ensure that companies follow the prescribed CSR activities, and the state-owned companies are even required by law to produce CSR disclosures [86]. This type of forced disclosure lacks quality, and it is used by the firms to hide financial results at the same time. Meanwhile, the voluntary regimes of the CSR disclosures produce high-quality reports.

According to Wang and Chen [87], the managers of the state-owned enterprises are likely to be involved in EM, because they lack competition. Additionally, the monitoring system of these companies can easily be bypassed. Besides these issues, the manager’s entrenchment also prevails in these systems. Noronha et al. [59] demonstrated that entrenchment issues reinforce the opportunistic behavior of the management in state-owned enterprises. Because they are state-owned, they are more focused on fulfilling the legal requirements, and they can easily conduct EM when they increase CSR-related practices.

We also contend that the Chinese state-owned organizations are more inclined to use REM over AEM because of the acceptable risk connected with AEM is comparatively greater with the state-owned firms compared to the non-state owned firms, which was mentioned by Kuo et al. [23] who held this conclusion after SSSREF (split share structure reform). Ewert and Wagen hofer [88] had earlier proposed a similar notion that was further endorsed by [56], but it is in contradiction to the findings of [22]. Thus, our second hypothesis is as follows.

H2. 

The state controlled firms moderate the relationship between CSR and EM.

3. Research Methodology and Sample

3.1. Sample and Data

The sample taken for this study consists of the companies that are listed on the Shanghai stock exchange and the Shenzhen stock exchange from 2010 to 2019. The unbalanced observations were 5457. The financial data for these companies was obtained from the CSMAR database. For the CSR rankings, we followed the rankings from the RKS database.

Table 1. Sample distribution.

Industry	Code	Observations
Agriculture	A	221
Mining & Dressing	B	283
Manufacturing	C	2328
Electric power & Supplies	D	307
Engineering Construction	E	229
Wholesale & Retail	F	301
Transportation	G	334
I.T	I	265
Real estate	K	361
Business services	L	202
Public facilities management	N	201
Journalism & publication	R	205
Conglomerate	S	220
Total		5457
Total number of companies	654

represents the sample detail.

3.2. Measurement of AEM

Since EM is conducted by manipulating the accounting choices, we used a modified version of the Dechow, Sloan, and Sweenedy (1996) [89] model as a proxy of AEM. By subtracting the nondiscretionary accruals from the total accruals, which is the difference between the observed and predicted y values, the error term is obtained. The error term represents the ha value of the discretionary accruals of the firm [90,91].

$$\frac{T{A}_{it}}{ASST{S}_{it-1}}={\beta }_0+{\beta }_1\frac{1}{ASST{S}_{it-1}}+{\beta }_2\frac{\left(\Delta {Sales}_{it}+\Delta {RCV}_{it}\right)}{ASST{S}_{it-1}}+{\beta }_3\frac{PP{E}_{it}}{ASST{S}_{it-1}}+{\in }_{it}$$

(1)

where TA_it represents the total accruals, ASSTS_it−1 represents the lagged of the total assets, β₀, β₁, β₂, and β₃ are equal to the coefficients of the model for the nondiscretionary accruals, ΔSales_it represents the current year sales less the previous years sales, ΔRCV_it represents the current years receivables less the previous years receivables, PPE_i represents plant, property, and equipment, and ∈_it represent the error term in year t for firm i.

3.3. Measurement of REM

We incorporated the different EM models in addition to those previously mentioned, so the results can be robust. These measures were based upon the studies by Roychowdhury [20] and Cohen and Zarowin [56], and they relate to the actual relationship between EM and the operating expenses. The normal level of the cash flows from operations, production costs, and other operating expenses were determined by subtracting them from the above normal values. We then used the ACFO, APC, and ADE variables as proxies for the REM.

$$\frac{CF{O}_{it}}{ASST{S}_{it-1}}={\beta }_0+{\beta }_1\frac{1}{ASST{S}_{it-1}}+{\beta }_2\frac{RE{V}_{it}}{ASST{S}_{it-1}}+{\beta }_3\frac{\Delta {REV}_{it}}{ASST{S}_{it-1}}+{\in }_{it}$$

(2)

One of the ways to manage earnings is to indulge in overproduction, and the production cost will tend to decline. This can be determined by the model cross-sectional regression for each industry and year.

$$\frac{PRO{D}_{it}}{ASST{S}_{it-1}}={\beta }_0+{\beta }_1\frac{1}{ASST{S}_{it-1}}+{\beta }_2\frac{RE{V}_{it}}{ASST{S}_{it-1}}+{\beta }_3\frac{\Delta {REV}_{it}}{ASST{S}_{it-1}}+{\beta }_4\frac{\Delta {REV}_{it-1}}{ASST{S}_{it-1}}+{\in }_{it}$$

(3)

The third type of REM relates to the cutting down of the discretionary expenses, such as marketing, research, and development. These actions are taken to artificially inflate the overall profits but have long-term detrimental effects on the companies [92]. In order to estimate these, we used the procedure identified by Roychowdhury [20].

$$\frac{Disc\_Ex{p}_{it}}{ASST{S}_{it-1}}={\beta }_0+{\beta }_1\frac{1}{ASST{S}_{it-1}}+{\beta }_2\frac{Sale{s}_{it-1}}{ASST{S}_{it-1}}+{\in }_{it}$$

(4)

The variable of the REM was constructed as per the suggestions from Roychowdhury [20] and Cohen and Zarowin [56]. They stated that the increase in activities raises the production costs and decreases the cash flows from operations, and it is measured using the following equation.

REM = ACFO − APC + ADE

(5)

where CFO represents cash flow from operations, PROD represents the sum of the cost of goods sold and the change in inventories, and Disc_Exp represent the discretionary expenses.

3.4. Measurement of Corporate Social Responsibility (CSR)

We relied upon the CSR provided by the Rankins CSR rating, which is commonly known as RKS. RKS was founded in 2002, and it is considered a reliable source of information. The index formulated by the RKS follows the guidelines set by the GRI (Global Reporting Initiative) in their report (GRI, 2002) along with criterion set by SA 8000 (SAI, 2001).

3.5. Measurement of State-Owned Enterprises (SOE)

As stated in various ownership studies [35,44,45,60,93,94,95,96,97,98,99,100,101,102], the control of the firm was determined by the pattern of shareholdings, for example if the majority of the shares are held by the government, then the company is considered as a state-owned enterprise (SOE). Otherwise, it will be considered as a private firm. The CSMAR database was utilized to determine this, because it is one of the most authentic sources of information when it comes to Chinese companies.

3.6. Control Variables

The control variables were chosen based upon the suggestions from the previous studies [21,22,23]. The control variables chosen were the size of the firm denoted by the F.Size, which was determined by taking the natural logs of the total assets of the firms. The debt to equity ratio or the leverage is denoted by Lev, which indicates the risk faced by the firms. Big 4 is the choice of audit firms (one if a firm’s auditor is one of the “Big4” accounting firms for the period and zero otherwise). Foreign institutional investors are denoted by FII, which is the presence of foreign ownership in companies listed in China, and is considered as a dummy variable in this research study. The independent directors are denoted by ID, which is calculated by the proportion of the independent directors divided by the total number of directors. The financial performance of the firms was represented by the returns on the assets, denoted by ROA.

Table 2. Definitions of variables.

Variable Name	Abbreviation	Description
Accrual earnings management	AEM	We used a modified Jones model as a proxy of AEM.
Real earnings management	REM	We used the Roychowdhry (2006) [20] model as proxy of REM.
Corporate social responsibility	CSR	We relied upon the CSR provided by the Rankins CSR rating, which is commonly known as RKS.
Firm Size	F.Size	Firm size was determined by taking the natural logs of the total assets of the firms.
Leverage	Lev	This is the total debt divided by total assets.
Big4	Big4	This value is 1if a firm’s auditor is one of the “Big4” accounting firms for the period, and 0otherwise.
Foreign institutional investors	FII	This represents the presence of foreign ownership in companies listed in China, and it is considered as a dummy variable in this research study.
State owned enterprises	SOE	If the majority of the shares are held by the government, then the company is considered as a state-owned enterprise (SOE). Otherwise, it will be considered as a private firm.
Independent director	ID	The proportion of independent directors divided the total number of directors.
Return on assets	ROA	This is calculated by dividing the net income by the total assets.

presents the definitions of variables.

3.7. Models

The association between the CSR and the EM can be mapped by the regression equations below.

AEM_it = β₀ + β₁CSR_it+ β₂CSR*SOE_it+β₃F.Size_it + β₄Lev_t+ β₅Big4_it+ β₆FII_it+ β₇SOE_it + β₈ID_it+ β₉ROA_it +β₁₀YearDum_it+ β₁₁Industry Dum_it +∈_it

(6)

REM_it = β₀ + β₁CSR_it+ β₂CSR*SOE_it+β₃F.Size_it + β₄Lev_t+ β₅Big4_it+ β₆FII_it+ β₇SOE_it + β₈ID_it+ β₉ROA_it +β₁₀YearDum_it+ β₁₁Industry Dum_it +∈_it

(7)

In the first equation, $AE{M}_{it}$ is the dependent variable, and $CS{R}_{it}$ is the independent variable.In the second equation, REM_it is the dependent variable, and CSR_it is the independent variable.

4. Results

4.1. Descriptive Statistics

The descriptive statistics obtained from the research sample are presented in

Table 3. Descriptive statistics.

Variable	Obs	Mean	Std. Dev
AEM	3.481	0.0764	0.0106
REM	3.481	−0.0997	0.0327
CSR	3.481	38.3450	9.1253
F.Size	3.481	24.2412	1.9690
Lev	3.481	0.5109	0.0140
Big4	3.481	0.1812	0.0852
FII	3.481	0.1559	0.0530
SOE	3.481	0.6217	0.1824
ID	3.481	0.3813	0.0757
ROA	3.481	0.0742	0.0264

Note: n = 5600. AEM refers to the accrual earnings management, and we used a modified version ofthemodel in [89] as a proxy. For REM, which refers to the real earnings management, we used the Roychowdhry (2006) [20] model as a proxy. CSR refer to corporate social responsibility. We relied upon the CSR provided by the Rankins CSR rating, which is commonly known as RKS. F.Size refers to the firm size, which is determined by taking the natural logs of the total assets of the firms. Lev, which refers to leverage, indicates the risk faced by the firms. The Big4 is the choice of audit firms, if the firm is audited by a “Big4” auditing firm for the period, Big4 is1; otherwise, it is 0. FII refers to foreign institutional investors, which is the presence of foreign ownership in companies listed in China, and it is considered as a dummy variable in this research study. ID refers to independent directors, which is calculated by the proportion of independent directors divided the total number of directors. ROA, which refers to the return on assets, is calculated by dividing the net income by the total assets.

. All the continuous variables in the question were winsorised (this is the transformation of the statistics by limiting the extreme values in the statistical data to reduce the effect of possibly spurious outliers) at both the 1st and 99th percentiles. The corresponding values of the CSR’s mean and standard deviation are 38.3450 and 9.1253. The mean value of the AEM is 0.0749, and the REM is −0.0997, which demonstrates that the AEM was favored over the REM by the Chinese companies that were included in the sample of the study. The mean values of SIZE is 24.2412, Lev is 0.5109, Big4 is 0.1812, and ROA is 0.0742.

4.2. Pearson Correlation

The values of Pearson’s correlation coefficients are presented in

Table 4. Correlation matrix.

	1	2	3	4	5	6	7	8	9
1-AEM	1
2-CSR	−0.013***	1
3-F.Size	0.069***	0.511***	1
4-Lev	0.093***	0.236***	0.518***	1
5-Big4	0.043**	0.426***	0.560***	0.232**	1
6-FII	0.037*	0.090***	0.155**	0.528**	0.181**	1
7-SOE	−0.238	0.130***	0.264***	0.199**	0.175**	0.089**	1
8-ID	−0.0672	0.018	0.054**	0.313**	0.061	0.030**	0.028**	1
9-ROA	−0.0418*	0.072	−0.137	0.377	0.215**	0.082**	0.117**	0.030	1

Note: n = 5600. AEM refers to the accrual earnings management, and we used a modified version of themodel in [89] as a proxy. For REM, which refers to the real earnings management, we used the Roychowdhry (2006) [20] model as a proxy. CSR refer to corporate social responsibility. We relied upon the CSR provided by the Rankins CSR rating, which is commonly known as RKS. F.Size refers to the firm size, which is determined by taking the natural logs of the total assets of the firms. Lev, which refers to leverage, indicates the risk faced by the firms. The Big4 is the choice of audit firms, if the firm is audited by a “Big4” auditing firm for the period, Big4 is 1; otherwise, it is 0. FII refers to foreign institutional investors, which is the presence of foreign ownership in companies listed in China, and it is considered as a dummy variable in this research study. ID refers to independent directors, which is calculated by the proportion of independent directors divided the total number of directors. ROA, which refers to the return on assets, is calculated by dividing the net income by the total assets. * p < 0.05, ** p < 0.01, *** p < 0.001.

along with their significance levels. The values indicate a negative correlation between CSR and AEM. As far as control variables are concerned, the results indicate that CSR is significantly positively correlated with F.Size, Lev, andBig4. The large-sized firms are more likely to be involved in CSR practices and vice versa. The maximum correlation is between the firm size and the Big4 and below the limit, so there is no problem of endogeneity.

4.3. Multivariate Analysis

Table 5. Regression results of CSR and EM.

Variables	Model 1 (AEM)		Model 2 (REM)
	Coef.	p-Value	Coef.	p-Value
CSR	−0.0219	0.026	−0.0113	0.051
F.Size	−0.0143	0.052	−0.0284	0.000
Lev	0.0540	0.000	0.0897	0.019
Big4	0.0541	0.395	0.0416	0.391
FII	−0.0104	0.072	−0.0256	0.158
SOE	−0.0162	0.016	−0.0108	0.044
ID	−0.0234	0.045	0.0405	0.067
ROA	0.0323	0.041	0.7179	0.000
Year & Ind	YES		YES
Adj−R2	25.1%		15.8%

Note: n = 5600. AEM refers to the accrual earnings management, and we used a modified version of themodel in [89] as a proxy. For REM, which refers to the real earnings management, we used the Roychowdhry (2006) [20] model as a proxy. CSR refer to corporate social responsibility. We relied upon the CSR provided by the Rankins CSR rating, which is commonly known as RKS. F.Size refers to the firm size, which is determined by taking the natural logs of the total assets of the firms. Lev, which refers to leverage, indicates the risk faced by the firms. The Big4 is the choice of audit firms, if the firm is audited by a “Big4” auditing firm for the period, Big4 is 1; otherwise, it is 0. FII refers to foreign institutional investors, which is the presence of foreign ownership in companies listed in China, and it is considered as a dummy variable in this research study. ID refers to independent directors, which is calculated by the proportion of independent directors divided the total number of directors. ROA, which refers to the return on assets, is calculated by dividing the net income by the total assets.

consists of the statistical results obtained from Model 1 and Model 2. The results from Model 1 indicate the adverse relationship between CSR and the discretionary accruals. AEM is considerably lower in firms with higher levels of CSR at a significance level of 95%. At the same time, CSR is also negatively correlated with REM in Model 2, with a p-value that is less than 0.05 and a coefficient value of −0.0113. The ethical viewpoint is supported by our results. The managers are least likely to engage in AEM if the firm increases the levels of CSR activities. In addition to this notion, it was also illustrated that REM is more sensitive to changes in CSR as compared to AEM. The transparency in terms of financial reporting in the case of the Chinese firms tend to improve when the companies raise the level of their CSR and ethical activities. Based on our results, we accept Hypothesis 1, and we also further endorse findings of [3].

When we analyzed the control variable of absolute discretionary accruals, it indicates a decreasing trend, but shows increasing trends in profitability and leverage. This can be explained by the agency cost theory and the political cost theory, because they indicate that larger firms with higher profit levels would tend to manage earnings to bring down the reported profits, which is suggested in [103]. In the present study, we found that the level of EM was associated with the size of the companies, because the larger firms have a lower amount of less discretionary accruals.

The leverage represented by Lev was considered as a control variable, and it was found to be positively correlated with the absolute discretionary accruals. Press and Weintrop [104] had earlier explained that higher levels of leverage are generally considered as a violation of debt convents, and DeFond and Jiambalvo [105] stated that the companies that are involved in these types of practices are also likely to use discretionary accrual to show better financial performances. In these types of cases, managers are always under pressure to enhance their financial performance using discretionary accruals. Lev is found to be positively related to AEM and REM, and the relationship is also significant. As mentioned above, the level of leverage is also indicative of the possibility of the EM. Meanwhile, the size of the firms also indicates the method of EM they are likely to use: Larger firms prefer to use AEM. Firms with high profitability, as indicated by ROA, as compared to REM, are also more likely to use AEM.

4.4. Two-Stage Least Square (2SLS) Results of CSR and EM

Additionally, to deal with the possibility of an endogeneity problem, we used the 2SLS regression model [16,58,106]. The main reason behind this application is that the managers assess the level of EM and CSR at the same time, and the issues related to endogeneity need to be addressed using the above—mentioned approach.

The results for the 2SLS are presented in

Table 6. SLS results of CSR and EM.

Variables	Model 1 (AEM)		Model 2 (REM)
	Coef.	p-Value	Coef.	p-Value
CSR	−0.0181	0.056	−0.0615	0.052
F.Size	−0.0147	0.051	−0.0264	0.000
Lev	0.0543	0.000	0.0296	0.945
Big4	0.0533	0.413	0.0442	0.030
FII	0.0105	0.067	0.0255	0.159
SOE	−0.0623	0.167	0.0109	0.436
ID	−0.0237	0.044	0.0359	0.712
ROA	0.0325	0.408	0.7168	0.000
Year & Ind	YES		YES
Adj-R2	25.1%		15.8%

Note: n = 5600. AEM refers to the accrual earnings management, and we used a modified version of themodel in [89] as a proxy. For REM, which refers to the real earnings management, we used the Roychowdhry (2006) [20] model as a proxy. CSR refer to corporate social responsibility. We relied upon the CSR provided by the Rankins CSR rating, which is commonly known as RKS. F.Size refers to the firm size, which is determined by taking the natural logs of the total assets of the firms. Lev, which refers to leverage, indicates the risk faced by the firms. The Big4 is the choice of audit firms, if the firm is audited by a “Big4” auditing firm for the period, Big4 is 1; otherwise, it is 0. FII refers to foreign institutional investors, which is the presence of foreign ownership in companies listed in China, and it is considered as a dummy variable in this research study. ID refers to independent directors, which is calculated by the proportion of independent directors divided the total number of directors. ROA, which refers to the return on assets, is calculated by dividing the net income by the total assets.

. The main goal here is to attain estimators that are consistent rather than gauging a goodness of fit as suggested by [107]. These results provide further confirmation of the results held in Table 5. The results show that CSR and AEM have a negative relationship at the 95% confidence level in Model 1 (Table 6), and the CSR and REM in Model 2 have a P-value that is less than 0.05. The value of the coefficient is −0.0615, and this also points to a strengthening of the existing significant relationship between the CSR and the REM at the 95% confidence interval. Based upon our results, we deduced that an increase in AEM and REM will hurt CSR activities by Chinese firms and vice versa, which are in line with the findings in [21]. Consequently, an ethical perspective (Hypothesis 1) is reinforced.

4.5. Moderating Role State Ownership (SOE) on CSR and EM

The result in

Table 7. Moderating role of state-owned enterprise (SOE)on CSR and EM.

Variables	Model 1 (AEMZ)		Model 2 (REM)
	Coef.	p-Value	Coef.	p-Value
CSR	−0.0370	0.052	−0.0265	0.676
CSR*SOE	−0.0146	0.000	0.0124	0.349
F.Size	−0.0306	0.012	−0.0241	0.000
Lev	0.0540	0.000	0.0135	0.750
Big4	0.0549	0.398	0.0353	0.069
FII	−0.0101	0.083	−0.0202	0.024
SOE
ID	−0.0120	0.069	−0.0683	0.094
ROA	0.0741	0.069	0.6691	0.000
Year & Ind	YES		YES
Adj−R2	11.9%		16.1%

Note: n = 5600. AEM refers to the accrual earnings management, and we used a modified version of themodel in [89] as a proxy. For REM, which refers to the real earnings management, we used the Roychowdhry (2006) [20] model as a proxy. CSR refer to corporate social responsibility. We relied upon the CSR provided by the Rankins CSR rating, which is commonly known as RKS. F.Size refers to the firm size, which is determined by taking the natural logs of the total assets of the firms. Lev, which refers to leverage, indicates the risk faced by the firms. The Big4 is the choice of audit firms, if the firm is audited by a “Big4” auditing firm for the period, Big4 is 1; otherwise, it is 0. FII refers to foreign institutional investors, which is the presence of foreign ownership in companies listed in China, and it is considered as a dummy variable in this research study. ID refers to independent directors, which is calculated by the proportion of independent directors divided the total number of directors. ROA, which refers to the return on assets, is calculated by dividing the net income by the total assets.

(Model 1) shows that the moderating variable (CSR*SOE) has a negative relationship with AEM, but the moderating variable (CSR*SOE) seems to lack any relationship with AEM, which does not come as a surprise as it was expected. This is indicative of the fact that the state-owned enterprises are much more likely to indulge in EM via REM when they decide to increase CSR activities. These results are supportive of the opportunism by the management of these enterprises. We find that the enterprises are more likely to favor REM over AEM as they increase the CSR activities, and these findings are in line with those held by earlier researchers (Ewert and Wagenhofer [88]; Cohen and Dey [108]). However, other researchers contradict the findings [22]. We can safely state that we accept our second hypothesis.

4.6. Discussion of Results

To date, the research on the topics of CSR has yielded mixed results. Some researchers hold that it is very important and state that it is a vital part of the overall long-term vision and strategy of any business, while others have linked it to the opportunistic behavior of management. One of the underlying theories that tend to explain why companies involve themselves in CSR activities is the stakeholder theory. This theory states that companies are aware of their stakeholder’s requirements and that, by involving themselves in these types of social activities, i.e. ethical behavior, they tend to increase their acceptability in wider society, which helps them to achieve their long-term goals and objectives. By conducting CSR activities and ethical behaviour, these companies are also able to incorporate the knowledge obtained from these activities into their future corporate strategies. Ullmann [109] had stated that the stakeholder theory forms the bases of the future success of any commercial organization, and the authors of the subsequently published [110] held similar views.

There is consensus amongst the researchers and academics that such programs are only successful if they are in line with the strategic vision and objectives of the companies. Along with that, the social activities selected for such programs and their potential impact on society or the environment needs to be communicated to such volunteers. This not only motivates the employees involved in the activity but also improves their commitment towards the cause and the organization as a whole. According to Brewis [7], such exercises not only improve employee satisfaction but also results in a positive influence that reaffirms employee loyalty, as it aligns with that of the organization. He further states that this also has a positive impact on the overall corporate image of the organization in the eyes of both employees and society as a whole.

There is a strong positive relationship between employee satisfaction and a positive attitude towards one’s job, as employee satisfaction tends to increase the overall morale of the organization. Tracey and Phillips [111] have also backed such claims and state that CSR activities and employee satisfaction are positively correlated. Caudron [112] calls this phenomenon “corporate citizenship” and states that this leads to an increase in loyalty from employees towards the organization and at the same time reduces employee turnover. Miller [113] also backs these findings and adds that this also improves the overall quality of the human resources of organizations.

The main aim of this research is to explore the association between CSR and EM practices in the Chinese context and to find out the moderating role of SOEs. The results of this research have led us to conclude the following when it comes to Chinese firms. The relationship between CSR and EM is negative, which supports the ethical viewpoint. Thus, it can be deduced that these firms are less likely to engage in EM practices when they increase their CSR activities, and these results are backed by earlier findings of Kim and Park [21] and those of [1].

Based upon our results, we further conclude that Chinese state-controlled firms are more likely to indulge themselves in EM practices, because they increase spending on CSR activities. Further investigation on the matter has revealed that these firms prefer REM when it comes to EM, because it is less detectable compared to other methods. In other words, they prefer to make changes in operating activities, such as policies for reconditioning income, manipulating the production level, and reconditioning discretionary expenses, because they increase their CSR activities. Our results are to a great extent in line with those held by Yip and Van Staden [114], which were later endorsed in [58]. According to the results, we can say that the state-controlled firms in China are more likely to be involved in EM.

5. Conclusions and Future Research

For the current research, the sample consisted of 5457 Chinese firm-year observation data on CSR activities from 2010 to 2019. We considered two main methods to measure the levels of EM: AEM and REM. Through our research, we identified the state control of firms as one of the major institutional factors moderating the relationship between CSR and EM. In the first step of our research, we found that the relationship between CSR and EM was negative. Afterwards, we explored the impact of the moderating variable, such as state control, on the above-mentioned relationship. To address the possible effects of endogeneity, the regression approach of 2SLS was adopted to test our research models.

To satisfy the stakeholders both internal and external, organizations need to incorporate CSR- and ethical-behavior-based policies not only into their operating environment but also into their organization’s culture. This view is backed by [65], the authors of which hold that employees are major (internal) stakeholders of any organization. As employees are major stakeholders, they also demand that the organization pay attention to CSR activities, i.e. ethical behavior, and when such expectations are fulfilled by organizations, the overall morale of the employees becomes satisfactory. This leads to an improvement in organizational commitment, a reduction in turnover, and an increase in productivity [115,116].

Our results confirmed the theoretical connections established earlier. We found that there was a negative relationship between CSR activities and EM. This means that, when Chinese firms increased their CSR activities, they tended to be less involved in EM activities. On the other hand, we also discovered that state-owned firms were more likely to increase REM while increasing CSR activities, because they were more likely to face increased legitimacy risk. At the same time, the possible incentives for potential EM are greater compared to the private sector companies [3]. In line with this notion, Shleifer [4] stated that companies with higher CSR activities are less likely to be involved in EM, because the general public and other stakeholders are more interested in their financial results, and the incentives for managers to practice unethical practices are usually lower.

These firms always prefer to benefit from these types of social relationships in the long term, and they ignore the short-term gains from indulging in EM activities. The quality of the earnings is preferred by these firms, because they are always under the public focus and other stakeholders. Researchers such as Orlitzky and Schmidt [9] and Brammer and Millington [117] have held that CSR and profitability are positively linked. Francoeur and Labelle [118] added that companies with a high level of CSR activities usually have a higher quality of earnings, compared to those that have a lower level of CSR activities.

The present research is helpful for both Chinese regulators and firms alike to understand the importance of CSR practices. The findings are important in the context of international and local investors along with academics, because they indicate the importance of CSR practices in the context of EM. This study can also be seen as a benchmark for other Asian markets that share similar characteristics with the Chinese markets [14,15]. Most of the Asian financial markets are considered as emerging markets, and they are of great importance for both local and international investors.

Nowadays, every organization wants to achieve a competitive advantage over its competitors and because of the additional financial resources required, achieving such an advantage isvery limited and hard to obtain. One way of achieving this is by having well trained and motivated employees. As most of the companies within a specific sector and category have similar financial resources, the human resource is the main defining factor in achieving this advantage [119]. Organizations can only achieve this if they cultivate a CSR- and ethics-based corporate culture, as only then can they realize the synergistic benefits. Gazzola [120] held that such a culture cannot be developed without the participation of employee satisfaction. The human resources department needs to take a lead role in encouraging employee participation in CSR- and ethics-based activities at all levels. This improves the working environment and culture of the organization, resulting in the positive work and ethical behaviour of employees ensuring long-term success for both the employees and the organization.

CSR and ethical behaviour not only have a positive impact on employee satisfaction but also improve investor confidence. Therefore, it has become vital for the organization to manage their volunteers and their CSR- and ethics-based projects in such a way that they not only help to improve an organization’s image but also prove to be a motivating and healthy activity for employee satisfaction.

Even though this study provides deep insight into the relationship between CSR and EM, it has a few limitations, because the study only encompasses Chinese listed firms that disclosed CSR practices and other information, such as control structure. The limitation mainly exists because many Chinese state-owned corporations are not even listed on the stock exchanges [121]. At the same time, we varied the types of institutional environments, which might greatly differ from one another depending upon the nature of their operations and the amount of information they tend to disclose. It is very difficult to gauge the level of EM in those enterprises. Many works in the past on topics such as Islam had have also argued about the impact of international markets on CSR practices in Chinese companies. Future studies can consider the present study as a foundation to investigate phenomena in Chinese financial markets. Any China-based international company may be more concerned about the image it portrays to its international investors and thus would extend its CSR practices, because they are of major concern for investors based in developed countries as compared to the local Chinese companies. In the future, researchers can compare the local CSR practices of local and international Chinese companies to validate or reject this type of hypothesis. We also conclude that there are issues of transparency when it comes to CSR practices and reporting with Chinese companies in general, even though China has made great progress in regard to the implementation of CSR reporting. These transparencies might be better than other developing countries, but they still significantly lag behind the developed markets.

This study provides useful insight into the CSR practices and their impact in terms of developing markets, but the researchers have to be careful with generalizing this study for all developing markets, because they might have different legal institutional requirements. Even though many Asian countries share a border with China, their legal and corporate sectors are much different than those of China, which is a result of the differences in culture, history, and colonial backgrounds.