Related Party Transactions and Earnings Management: The Moderating Effect of ESG Performance

Abstract

The purpose of this paper is to investigate the effect of ESG performance on the possible association between related party transactions (RPTs) and different types of earnings management (EM). We study a sample of Italian non-financial listed firms over the 2014–2019 period, controlling for the fixed effects of the company’s sector of operation and the year. We investigate the effect of Environmental, Social and Governance (ESG) performance scores on the association between related party transactions and real earnings management (REM), as well as accrual-based earnings management (AEM). We provide evidence that firms might use RPTs in association with downward AEM or as a substitute of REM perpetrated by sales anticipations and discretionary expenses reductions, as well as an autonomous form of earnings manipulation. Our empirical evidence shows a significant moderating effect of ESG performance on earnings management. In particular, social as well as governance performance significantly moderate the association between RPTs and downward AEM; environmental performance moderates the possible use of RPTs in association with cashflow-based REM. This is the first study that analyzes the effect of ESG performance on the possible association between related party transactions and earnings management.

1. Introduction

Literature points out that many financial scandals that have involved famous companies such as Enron, Adelphia, Worldcom, Subprime Mortgage in the USA, Bank of Credit and Commerce in UK, Parmalat and Cirio in Italy are related to earnings manipulations and fraudulent related party transactions [1,2].

Accounting scandals in large firms have raised concerns about the quality of financial information as well as the interest of scholars for its various aspects, among them earnings management (EM). Managers can practice accrual earnings management (AEM) by exploiting the characteristics of the accrual accounting system [3]. They can carry out real earnings management (REM), departing from normal business operations—accelerating sales, offering price discounts, reducing discretionary expenditure, delaying research and development and maintenance expenditures—in order to mislead stakeholders into thinking that financial results were produced by normal operational practices [4]. They can also boost earnings [5,6] by means of related party transactions (RPTs) between a group’s companies or use RPTs in order to obtain benefits or to direct profits to controlling shareholders [7].

Together with the concern for the quality of financial reporting, the interest of investors and stakeholders, in general, for corporate non-financial engagement and performance has grown. Literature has widely addressed the issue of the connections between corporate social responsibility (CSR) and EM, pointing out a significant relationship among others [8,9,10,11,12,13,14,15]. A limited number of them used ESG performance indicators to investigate the association between ESG performance and EM [16,17,18], pointing out a negative relation. Recent literature calls for the further investigation of the possible moderating effect of ESG factors on earnings quality [19].

Fewer scholars have investigated the relation between RPTs and EM [2,20,21,22], but results are quite mixed, highlighting that this issue deserves further study. This stream of literature has also taken into consideration the effect of ownership concentration [20], ownership type [22] and of auditor type [21]. However, to the best of our knowledge, no study has investigated the effect of CSR/ESG on the possible relation between RPTs and EM.

Given this literature gap, the research question is as follows: “Are abnormal RPTs associated to earnings management? If so, does ESG performance moderate the relationship?”.

This paper draws on the Agency Theory and the Stakeholder Theory as its aim is to analyze how agency conflicts might combine RPTs with EM, verifying the possible moderating effect of environmental, social and governance performance.

This study addresses the research question by focusing on a sample of 76 Italian non-financial listed firms for the period 2014–2019. Italian financial market is characterized by high ownership concentration, with limited possibility to contest the control [23] (p. 13). In 2020, the ownership stake held by the largest shareholder was, on average, 47.6% [24]. This scenario increases the likelihood of agency conflicts between minority and majority shareholders. Since RPTs raise concerns for investor protection, in 2010 the Italian securities regulator (Commissione Nazionale per le Società e la Borsa, CONSOB) adopted the “RPTs Regulation”, with Resolution no. 17221/2010. The RPTs Regulation sets forth provisions aimed at safeguarding transparency as well as procedural and substantive fairness of RPTs. The Italian context is of interest as, although RPTs are strictly regulated, they have nonetheless been used as opportunistic tools and have led to large-scale financial scandals [2]. Italy is also of interest given the recent promotion of non-financial information disclosure through Legislative Decree No. 254/2016, which implemented Directive 2014/95/EU in Italy. According to it, listed companies, with more than 500 employees, have been required to present non-financial information starting from 2017. This type of disclosure encompasses environmental issues, social and employee-related matters, human rights, diversity and anti-corruption.

This paper studies the association between RPTs and different forms of real as well as accrual-based earnings management, analyzing the possible moderating effect of environmental, social and governance performance.

Given the moderating effect of corporate governance characteristics on the relation between RPTs and EM [2,20,22], it controls for the effect of certain board characteristics, a company’s ownership type and some firm’s characteristics.

Results indicate a significant negative relation between abnormal RPTs and REM via abnormal costs of production, a significant positive relation between abnormal RPTs and downward accrual earnings management and real earnings management perpetrated by abnormal levels of discretional expenses. This study contributes to the field of study on EM and RPTs as, according to Agency theory, results suggest that abnormal levels of RPTs are used as a substitute, or in association with, different types of earnings management.

Moreover, consistent with Stakeholder theory and the legitimacy perspective, findings point out that social and governance performance moderates the association between abnormal RPTs and downward accrual earnings management whilst environmental performance moderates the relation between RPTs and REM perpetrated by cash-flow manipulations. This research contributes to the debate on the relation between ESG/CSR and EM as it provides evidence, for the first time, of the moderating effect of ESG performance on the relation between RPTs and EM.

The remainder of the paper is organized as follows: Section 2 introduces the theoretical framework, the literary review and hypotheses development; Section 3 provides the data and methods; Section 4 contains the results and the discussion; Section 5 concludes by pointing out contributions, implications and limits of the study as well as avenues for further research.

2. Theoretical Framework, Literature Review and Hypotheses Development

2.1. Theoretical Framework

According to Agency theory, the separation of a firm’s ownership and control results in agency conflict and information asymmetry between the owner of the capital and the management [25]. Owing to information asymmetry, earnings management activities may occur when directors are motivated to pursue their self-interest at the expense of stockholders’ interests [26]. They may manipulate to increase a firm’s performance in order to increase the related managerial benefits or, if the annual benchmark has been reached, to reduce the current year’s economic result in order to shift profits to the future and guarantee benefits for the following year [27]. A second type of agency conflict occurs between majority and minority shareholders [28] and may result in downward earnings management in order to reduce dividends, increase self-financing and prevent the use of equity financing and control dilution [29]. This type of agency conflict may also motivate transactions with related parties aimed at transferring resources where controlling shareholders have higher cash flow rights [30].

According to Stakeholder theory, a company should seek to meet the expectations of a broad range of constituencies, defined as “a person or group that can affect or is affected by the achievement of the organization’s objectives” [31] (p. 46). The concept of stakeholders relates to the inter-relationship between the company and different groups of individuals, i.e., shareholders, employees, suppliers, customers, environment and community. According to this theory, firms should adapt their actions to the values of their stakeholders in order to meet their expectations [32]. Stakeholder theory conceptualizes the firm as a part of a broader social system and it extends the management’s focus beyond maximizing the interests of stockholders to include other groups’ interests that may be affected by a company’s activities and, in turn, affect its behavior [33]. Agency and stakeholder theories complement each other by supporting the alignment of stockholder, stakeholder and management goals [34]. Both frameworks discourage the opportunistic behavior of management [35]. Agency theory states that earnings management causes agency problems between managers and shareholders [36]. Consistently, according to Stakeholder theory, managerial opportunism, earnings manipulations and any socially unacceptable strategies can harm a firm’s relations with its stakeholders [37,38]. From this point of view, the disclosure of financial, as well as social and environmental information, affects the relationship between a company and its stakeholders as information on a company’s activities legitimizes its behavior and changes stakeholders’ perceptions and expectations [39,40].

On the one hand, CSR performance produces legitimacy [41]; on the other, an increase in a company’s engagement in CSR can enhance stakeholders’ expectations and the pressure for responsible behavior in other business practices too [42].

Legitimacy is the perception that a company’s activities are desirable, appropriate and reflect socially developed norms, values, beliefs and definitions [43]. When a company does not operate in accordance with society’s norms and expectations, a legitimacy gap emerges that eventually leads to the withdrawal of the company’s legitimacy, thus threatening its survival [44]. Stakeholders concede pragmatic legitimacy to a company as long as they perceive that they will benefit from its activities, as well as moral legitimacy when company’s goals are pursued in a socially acceptable manner [45].

2.2. Earnings Management and RPTs

Literature has offered two different views on the nature of RPTs. On the one hand, research [46] suggests that intercompany transactions between companies within business groups might enhance a firm’s efficiency as it may obtain financial support from other firms in the group without resorting to external financing. According to existing literature, this kind of RPTs refers to the efficient transaction hypothesis [47,48]. Under this view, RPTs may be an effective method to reduce transaction costs through the creation of a corporate group internal market [6].

On the other hand, the literature indicates that managers might use related party transactions in order to obtain benefits or to transfer profit and assets out of the firm in order to maximize majority shareholder benefits to the detriment of minority shareholders [7]. This kind of RPTs corresponds to the conflict of interest’s transaction hypothesis [48], and it damages corporate value [49]. When these type of transactions occur, a firm’s management has incentives to use earnings management practices to justify, or increase, the benefits extracted or to mask the expropriation of resources in favor of controlling shareholders [2].

Cheung et al. [50] found that companies pay a higher price or receive a lower price compared to similar arms-length deals, respectively, in acquiring or in selling assets to related parties. There is also evidence that RPTs are commonly used in order to boost earnings [6] and other authors suggest that most offshore-related party transactions are arranged for this purpose [51]. Consistently, research has found that many companies use RPTs to enhance their performance during the pre-initial public offering (IPO) period [5], nevertheless the decrease in operating RPTs after IPO negatively affects a company’s stock return and contributes to reducing its post-IPO long-term performance [52].

There is evidence that the controlling shareholder has the ability to manipulate participations in a group’s firms to obtain personal benefits through RPTs [53]. Consistently, empirical evidence [30] has also pointed out that business group owners are prone to move profits from a company where they have low cash flow rights to a company where they have high cash flow rights.

Munir et al. [20] analyzed the effect of RPTs on earnings management, finding that tunneling practices are positively associated to earnings management for family firms with highly concentrated ownership. Furthermore, there is evidence that corporate governance quality lessens the positive association between sales-related party transactions and earnings management [2].

Literature has also investigated whether RPTs are associated to real or accrual earnings management or used as a third alternative to manage reported earnings [21]. Findings show that, on average, if a company’s auditor is not one of the ‘Big Four’, real earnings management and RPTs are negatively related and that there is not a significant relation between related party transactions and accrual earnings management. Subsequent research has analyzed the relation between EM and RPTs, taking into account the effect of ownership structure. Results pointed out a negative relation between RPTs and AEM and that the presence of higher levels of institutional ownership are positively related to the tendency to use RPTs in association with REM [22].

Empirical research, overall, suggests that companies might use RPTs as an earnings management practice. Moreover, empirical study has provided evidence that companies make choices between real earnings management and accrual earnings management [54,55], pointing out that when accrual-based earnings management becomes costly, managers shift to real earnings management, and vice-versa [56]. Therefore, we posit the following.

Hypothesis 1a (H1a).

Abnormal related party transactions are associated to real earnings management.

Hypothesis 1b (H1b).

Abnormal related party transactions are associated to accrual earnings management.

2.3. Earnings Management and ESG

A number of studies have addressed the issue of the relationship between CSR and earnings management practices. A majority of studies have reported a negative relation although a few studies found a positive relation, or a negative relation, for a specific CSR component or different results according to the type of earnings management taken into account.

Some studies have reported a negative link between CSR and earnings quality [57,58], suggesting that firms that engage in poor accounting practices try to compensate stakeholders by employing CSR [59]. Other studies have investigated the use of CSR and its disclosure for camouflage purposes. They found evidence of these practices only for family firms [60] and in cases of managerial entrenchment, resulting in a gap between CSR disclosure and CSR performance [61].

On the other hand, Chih et al. [62] investigated the relationship between CSR and earnings management and found that more socially and environmentally responsible companies tend to smooth earnings less and engage less in avoiding earnings losses and decreases but are prone to earnings aggressiveness. Nevertheless, research focusing on the effect of environmental performance on earnings management practices concluded that the former reduces accrual earnings management whilst it increases real earnings management [19].

Choi and Pae [63], analyzing the relationship between business ethics and financial reporting quality, found that companies with a high ethical commitment present a higher financial reporting quality and engage in less earnings management. Research also provided evidence that government-mandated CSR policy engenders conservative financial reporting [12].

Kim et al. [37], by controlling for different CSR dimensions and components, pointed out that socially responsible firms are less likely to resort to accrual or real earnings management and are less likely to be subject to SEC investigation, supporting the view that CSR reporting and performance are positive reputation signals related to lower earnings management practices [64]. The negative relation between CSR activities and real earnings management practices has been reported by other studies [8,15,16,17] also pointing out the moderating effect of corporate governance characteristics [11].

The effect of a company’s CSR orientation on its reporting incentives has also been studied in terms of the trade-off between REM and AEM [10], providing evidence that CSR-oriented companies are less likely to engage in REM because of its detrimental value on future performance. In addition, CSR may act as a concealer [61].

In common with the present study, recent empirical literature has focused on the relation between ESG performance and earnings manipulations in terms of financial irregularities, accrual and real earnings management, but unlike this work, no research has addressed the effect of ESG performance on the association between RPTs and EM. There is evidence that ESG performance mitigates financial irregularities [65], tax avoidance practices [66] and accrual-based [67] as well as real earnings management [18], particularly in countries where integrated reporting is mandatory [14].

ESG performance represents a company’s “configuration of principles of environmental, social and governance responsibility; processes of environmental, social and governance responsiveness, and politics, programs and observable outcomes as they relate to the firm’s societal relationships” [16] (p. 322). A company’s focus on ESG and sustainable business operations suggests that management seeks to make responsible decisions by taking care of stakeholders’ interests and expectations, whilst earnings management practices reveal the opposite behavior in conducting business.

Research has found a negative relation between ESG and earnings management practices and a positive relation between ESG and financial reporting quality, in line with Stakeholder theory. This suggests that the concern for ESG is a form of management responsibility that can reduce agency costs.

ESG performance increases stakeholder confidence that a firm’s behavior is consistent with their expectations [68]. Furthermore, ESG performance can enhance stakeholders’ expectations and, in turn, the pressure on management to employ ethical business practices [42] in order to preserve a firm’s legitimacy in the eyes of its stakeholders.

Therefore, this study posits the following.

Hypothesis 2a (H2a).

ESG performance moderates the relation between related party transactions and real earnings management.

Hypothesis 2b (H2b).

ESG performance moderates the relation between related party transactions and accrual earnings management.

3. Data and Methods

3.1. Data

The initial data sample comprised the population of non-financial firms listed on the Italian stock exchange in 2019, removing insurance and financial firms, given their accounting and regulatory peculiarities. The analysis covers the period 2014–2019.

The sample of Italian non-financial listed firms with available information on related party transactions and ESG scores comprises 76 firms as of 2019. Data on RPTs were hand-collected from the notes to the consolidated financial statements. Environmental, Social and Governance scores were collected from the Refinitiv-Eikon Database [69]. Financial and accounting data were collected from Orbis, the global Bureau van Dijk database, using consolidated financial statements. Board attributes were reconstructed based on the information available on the Orbis database and on the Chamber of Commerce register.

3.2. Dependent Variables

Firms may use several accrual and real earnings management strategies [70,71]. Following prior literature [4,21,71] this study proxies accrual-based earnings management with discretionary accruals and real earnings management with abnormal levels of cash flow from operations (RMCFO), abnormal levels of production costs (RMPROD) and abnormal levels of discretionary expenses (RMDISX).

Discretionary accruals (DA) for firm i in period t are the difference between total accruals and non-discretionary accruals. Dechow et al. [72] suggest a relationship between accrual earnings management and financial performance and, following Kothari et al. [73], ROA is included in the following model:

TA_i,t/A_i,t−1 = α₀ + α₁(1/A_i,t−1) + α₂[(ΔREV_i,t − ΔREC_i,t)/A_i,t−1] + α₃(PPE_i,t/A_i,t−1) + α₄(NI_i,t/A_i,t−1) + ε_i,t

(1a)

where

• A_i,t−1 = lagged total assets;
• ΔREV_i,t = change in revenues between period t and t − 1;
• ΔREC_i,t = change in trade receivables between period t and t − 1;
• PPE_i,t = gross property, plant and equipment;
• NI_i,t = net income.

Normal levels of cash flows from operations are estimated as in Dechow et al. [74].

CFO_i,t/A_i,t−1 = α₀ + α₁(1/A_i,t−1) + α₂(REV_i,t/A_i,t−1) + α₃(ΔREV_i,t/A_i,t−1) + ε_i,t

(1b)

Normal levels of production costs are estimated with the following model.

PROD_i,t/A_i,t−1 = α₀ + α₁(1/A_i,t−1) + α₂(REV_i,t/A_i,t−1) + α₃(ΔREV_i,t/A_i,t−1) + ε_i,t

(1c)

Normal levels of discretionary expenses (advertising, R&D and SG&A) are estimated in Equation (1d).

DISX_i,t/A_i,t−1 = α₀ + α₁(1/A_i,t−1) + α₂(REV_i,t/A_i,t−1) + α₃(ΔREV_i,t/A_i,t−1) + ε_i,t

(1d)

High values for the estimated residuals in the accruals and cost of production equations (low values in the cash flow and discretionary expenses equations) proxy for abnormal levels of earnings management indicating real activities manipulation and discretionary accruals (given the dimension of the data sample, Equation (1a,d) are estimated for the full sample).

3.3. Explanatory and Control Variables

The explanatory and control variables are ARPTs, the ESG scores ENVS, SOCS and GOVS, BSIZE, WBD, CEOD, DFF, ROA, SIZE, LEV, AGE, MKTBOOK, year and industry dummies (

Table 1. Description of variables.

Variable	Description	Measurement
NDA, PDA	Accrual earnings management	Negative (NDA) or positive (PDA) discretionary accruals using the Kothari model [73]
RMCFO	Real earnings management	Abnormal levels of cash flow from operations [4,21,74]
RMPROD	Real earnings management	Abnormal levels of production costs [4,21,74]
RMDISXS	Real earnings management	Abnormal levels of discretionary expenses [4,21,74]
ARPTs	Related party transactions	Related party transactions, measured as the abnormal level of the continuous variable RPTs/Assets
ENVS, SOCS, GOVS	ESG scores	Firm’s scores for the Environmental, Social and Governance pillars [69]
BSIZE	Governance	Number of directors in the board
WBD	Governance	weight of women that sit on the board of directors
CEOD	Governance	Dummy variable equal to 1 if the CEO is also the Chairperson and 0 otherwise
DFF	Governance	Dummy variable equal to 1if a family owns at least 20% of the firm’s common shares
SIZE	Other control variable	Log of firm’s total assets
ROA-1	Other control variable	Net income for year (t − 1) scaled by beginning of reporting period total assets
LEV	Other control variable	Sum of long term and short-term financial debts scaled by assets
MKTBK	Other control variable	Market value of common shares/Asset value
AGE	Other control variable	Firm’s age in years

).

RPTs are measured as the value of RPTs scaled by total assets [75]. RPTs is the sum of sales, purchases and outstanding balances between the company and its related parties [21,22]. Ryngaert and Thomas [76], using a sample of small-to-medium sized firms, suggest using a dummy variable equal to 1 if the sum of RPTs exceeds 1% of assets, but the choice of this cutoff point is arbitrary. They also indicate that their results are similar with alternative cutoffs and using the value of RPTs scaled by total assets. Given the characteristics of the sample, composed of listed firms, this analysis uses the RPTs-to-assets ratio, and, to reduce measurement errors in the model, it includes, as an explanatory variable, the abnormal (ARPTs) or estimated residual of the related party transactions variable derived from Equation (1e):

RPTs_i,t = α₀ + α₁(A_i,t) + α₂(MKTBOOK) + α₃(LEV)) + ∑α_kD_k + ε_i,t

(1e)

where

• MKTBOOK = market-to-book ration;
• LEV = financial debts scaled by assets;
• Dk = year and industry dummies.

The firm’s ESG variables are based on the relative performance of ESG factors with respect to the industrial sector and country of incorporation, producing for each firm a score between 0 and 100. The model incorporates the scores for the Environmental, Social and Governance pillars separately [69].

Control variables include governance and other firm characteristics that could affect the results. The governance variables are CEOD, BSIZE, WBD and DFF. CEOD is a dummy that receives the value of 1 if the same person is the CEO and the chairperson of the board of directors [77]. Board size (BSIZE) is the number of members on the board [2], and WBD is the weight of women that sit on the board of directors [78,79]. The model controls for the effect of family involvement on earnings management and related party transaction [7] including a dummy variable (DFF) that takes the value of 1 if a family owns at least 20% of the firm’s common shares.

The other firm characteristics included as control variables are SIZE, ROA, LEV, MKTBOOK and AGE and dummies to control for differences across industries and years. Firm size (SIZE) is proxied by the logarithm of assets [80,81]. The model also includes ROA to control for possible effects of this variable [4]. Financial leverage could be associated with real earnings management [82,83,84]. LEV is the sum of long-term and short-term financial debts scaled by assets. The model also includes MKTBOOK, the market-to-book ratio, as Dechow et al. [85] found that earnings manipulation could be related to abnormal market-to-book ratios. Finally, the model includes firm’s age (AGE) and dummies for year and industry effects as research points out that earnings management practices might vary across different stages of a firm’s life cycle [86,87] as well as across different sectors of activity [88].

3.4. Models

Having four proxies for earnings management activities, the model (Model 1) is described as follows:

EMi,t = β0 + β1RPTsi,t + ∑βjESGj,i,t + ∑βhXh,i,t + ∑βkXk,i,t + εi,t

(2)

where

• EMi,t = one of the four proxies of real and accrual earnings management;
• RPTsi,t = related party transactions for firm i in period t;
• ESGj,i,t = Environmental, Social and Governance scores;
• Xh,i,t = Governance control variables;
• Xk,i,t = Other control variables.

The following augmented model (Model 2) incorporates possible interaction effects between related party transactions and Environmental, Social and Governance scores:

EM i,t = β0 + β1RPTsi,t,+ ∑βjESGj,i,t + ∑βhXh,i,t + ∑βkXk,i,t + βw Interaction variable + εi,t

(3)

Earnings management activities and financial performance could be endogenously determined, and this may result in simultaneity, reversed causality or other endogeneity problems in the above model estimation. To deal with these concerns, a variety of econometric approaches has been proposed: fixed effects, lagged variables, control variables and generalized method of moments (GMM). Following Li [89], the analysis addressed these issues by using previous year financial performance (ROA-1) as a control variable to improve the robustness of the results.

The models are estimated with Generalized Least Squares (GLS) with clustering at the firm level, the standard errors are robust for the presence of correlations across firms. Year and industry dummies control for time and industry fixed effects.

The dependent and explanatory variables are detailed in Table 1.

4. Results

4.1. Descriptive Statistics

Panels A and B in

Table 2. Descriptive statistics for the whole sample and by industry.

A. Descriptive Statistics
		Mean			Median			Standard Deviation
NDA		−0.041			−0.026			0.045
PDA		0.049			0.024			0.060
RMCFO		0.000			0.014			0.063
RMPROD		0.000			0.026			0.193
RMDISX		0.000			0.010			0.145
RPTs		0.061			0.033			0.090
ABRPTs		−0.003			−0.010			0.087
BSIZE		11.527			11			3.773
ENVS		55.764			60.53			26.84
SOCS		65.806			68.95			20.21
GOVS		52.356			52.42			20.94
WBD		0.345			0.333			0.113
CEOD		0.265			-			-
DFF		0.50			-			-
ROA-1		0.078			0.066			0.075
SIZE		15.187			15.167			1.760
LEV		0.269			0.243			0.152
MKTBK		0.998			0.571			1.208
AGE		63.022			57			39.1
B. Descriptive Statistics by Industry
Industry	Weight	NDA	PDA	RMCFO	RMPROD	RMDISX	RPTs	ENVS	SOCS	GOV
Apparel and Textile	9.78%	−0.025	0.022	0.0002	0.0003	−0.0000	0.020	0.425	0.555	0.401
Chemical and Farmaceutical	8.44%	−0.036	0.070	0.0004	−0.0004	0.0011	0.022	0.447	0.555	0.470
Computer and Household	4.44%	−0.031	0.071	0.0004	0.0006	0.0003	0.023	0.592	0.709	0.353
Industr, Commercial and Transport Equipment	16.44%	−0.037	0.021	0.0002	0.0001	0.0002	0.055	0.592	0.712	0.528
Utilities and Related Services	22.22%	−0.033	0.052	0.0001	−0.0000	0.0002	0.077	0.686	0.679	0.568
Construction and Building Materials	9.78%	−0.039	0.052	0.0001	0.0003	−0.0006	0.142	0.582	0.685	0.590
Wholesale and Retail Trade	5.33%	−0.060	0.047	0.0001	−0.0001	−0.0002	0.041	0.480	0.615	0.510
Publishing, Motion and Entertainment	2.67%	−0.128	0.061	−0.0001	−0.0001	0.0008	0.065	0.345	0.581	0.410
Digital Solutions, Software and Consulting	18.22%	−0.048	0.075	0.0003	0.0005	0.0000	0.059	0.544	0.686	0.600
Real Estate	2.67%	−0.057	0.041	0.0002	0.0021	0.0001	0.028	0.427	0.640	0.408

present descriptive statistics for the variables in our sample. The mean values for the real earnings management proxies RMCFO, RMPROD and RMDISX are close to zero. It can be seen from the table that RPTs represent, on average, 6.1% of assets, but the median value is 3.3%. Board size shows a median value of 11 members, and women on board weigh on average 34.5%, while CEO duality indicates a 73.5% separation between CEO and chairperson. The Environmental, Social and Governance scores range, on average, from 52 to 66, leverage is 27% and the market-to-book ratio is, on average, close to 1.

The descriptive statistics in panel B show the distribution of the sample firms by industry and the disaggregation of the accrual and real earnings management proxies, related party transactions and ESG scores. The Construction industry has the highest value of related party transactions and the Apparel industry the lowest. ESG scores have values on average over the global sample mean in particular for the Utilities, Equipment and Computer industries and, conversely, values under the global mean in particular for the Publishing, Chemical and Apparel industries.

Table 3. Correlation matrix.

	PDA	RMCFO	RMPROD	RMDISX	ABRPTs	BSIZE	WBD	CEOD	DFF	ROA-1	SIZE	LEV	AGE	MKTBK	ENVS	SOCS	GOVS
NDA	-	−0.135	−0.177	0.051	0.204	−0.042	−0.050	−0.092	−0.143	−0.128	0.144	0.056	−0.049	0.012	0.038	0.039	0.024
PDA		0.160	0.035	0.109	−0.113	0.112	−0.090	−0.080	0.028	−0.356	−0.238	−0.023	−0.130	−0.089	−0.348	−0.277	−0.168
RMCFO			0.404	−0.130	0.132	0.111	−0.048	0.084	0.014	0.460	−0.079	−0.067	−0.119	0.553	−0.040	−0.014	−0.002
RMPROD				−0.700	−117	0.073	−0.059	0.238	0.148	0.326	−0.035	−0.021	−0.142	0.328	−0.051	−0.021	−0.017
RMDISX					0.181	−0.061	−0.004	−0.124	−0.069	−0.071	0.088	−0.003	0.131	−0.150	0.016	−0.002	−0.010
ABRPTs						−0.026	−0.036	−0.054	−0.104	0.051	0.078	−0.146	−0.185	0.071	0.014	0.017	−0.003
BSIZE							0.002	−0.209	0.087	0.171	−0.029	−0.021	0.224	0.125	0.160	0.162	0.148
WBD								−0.305	−0.289	0.026	0.039	0.088	0.117	0.006	0.037	0.020	0.058
CEOD									0.097	0.155	−0.226	0.076	−0.167	0.199	−0.057	−0.057	−0.078
DFF										0.100	−0.381	−0.206	0.021	0.180	−0.146	−0.110	−112
ROA-1											−0.176	−0.179	−0.061	0.689	−0.058	−0.043	−0.044
SIZE												0.130	0.258	−0.288	0.159	0.119	0.109
LEV													0.018	−0.284	0.083	0.058	0.077
AGE														−0.063	0.119	0.110	0.088
MKTBK															−0.082	−0.059	−0.057
ENVS																0.990	0.980
SOCS																	0.988

Bold indicates that the estimated correlation coefficient is significant at the 5% level.

shows that abnormal related party transactions (ARPTs) are correlated with negative discretionary accruals and with abnormal discretionary expenses. Abnormal production costs are negatively correlated with negative discretionary accruals. Positive discretionary accruals are negatively correlated with the environmental and social scores. The correlation matrix shows moderate to average associations between the variables, with the exception of ESG scores that are highly correlated with each other, and in order to avoid multicollinearity problems, they are used separately in regression models.

4.2. Empirical Results

Table 4. Earning management, related party transactions and ESG scores.

	Model 1		Model 1	Model 1		Model 1	Model 1		Model 1
	NDA		NDA	NDA		PDA	PDA		PDA
INT	−0.08(0.04) **		−0.08(0.04) **	−0.08(0.04) **		0.29(0.10) ***	0.33(0.11) ***		0.33(0.12) ***
ABRPTs	0.07(0.03) **		0.07(0.03) **	0.07(0.03) **		0.03(0.10)	−0.00(0.09)		−0.02(0.09)
BSIZE	−0.00(0.00)		−0.00(0.00)	−0.00(0.00)		0.00(0.00)	0.00(0.00)		0.00(0.00)
WBD	−0.03(0.04)		−0.03(0.04)	−0.03(0.04)		−0.10(0.07)	−0.10(0.07)		−0.10(0.07)
CEOD	−0.02(0.01)		−0.02(0.01)	−0.02(0.01)		−0.01(0.02)	−0.02(0.02)		−0.01(0.02)
DFF	−0.02(0.01) *		−0.02(0.01) *	−0.02(0.01) *		0.01(0.01)	0.01(0.01)		0.01(0.01)
ROA-1	−0.18(0.04) ***		−0.18(0.04) ***	−0.18(0.04) ***		−0.56(0.23) **	−0.59(0.23) **		−0.63(0.23) ***
SIZE	0.01(0.00) **		0.01(0.00) *	0.01(0.00) *		−0.01(0.01) *	−0.01(0.01) **		−0.02(0.01) **
LEV	0.01(0.04)		0.01(0.04)	0.01(0.04)		−0.04(0.05)	−0.04(0.06)		−0.04(0.06)
AGE	0.00(0.00)		0.00(0.00)	0.00(0.00)		−0.00(0.00)	-0.000(0.00)		−0.00(0.00)
MKTBK	0.01(0.00) ***		0.01(0.00) ***	0.01(0.00) ***		0.01(0.01)	0.01(0.01)		0.01(0.01)
ENVS	−0.00(0.00)					−0.00(0.00)
SOCS			−0.00(0.00)				−0.00(0.00)
GOVS				−0.00(0.00)					0.00(0.00)
INDUSTRY	yes		yes	yes		yes	yes		yes
YEAR	yes		yes	yes		yes	yes		yes
obs	113		113	113		93	93		93
R2	0.24		0.24	0.24		0.39	0.36		0.36
	Model 1	Model 1	Model 1	Model 1	Model 1	Model 1	Model 1	Model 1	Model 1
	RMCFO	RMCFO	RMCFO	RMPROD	RMPROD	RMPROD	RMDIX	RMDIX	RMDIX
INT	0.02(0.06)	0.03(0.06)	0.03(0.06)	−0.14(0.17)	−0.13(0.17)	−0.13(0.17)	0.05(0.19)	0.04(0.19)	0.04(0.19)
ABRPTs	0.06(0.10)	0.06(0.10)	0.06(0.10)	−0.27(0.10) ***	−0.27(0.10) ***	−0.27(0.10) ***	0.30(0.11) ***	0.30(0.11) ***	0.30(0.11) ***
BSIZE	−0.00(0.00)	−0.00(0.00)	−0.00(0.00)	0.00(0.00)	0.00(0.00)	0.00(0.00)	−0.00(0.00)	−0.00(0.00)	−0.00(0.00)
WBD	−0.08(0.04) *	−0.08(0.04) *	−0.08(0.04) *	−0.07(0.15)	−0.06(0.15)	−0.07(0.15)	−0.02(0.14)	−0.02(0.14)	−0.02(0.14)
CEOD	−0.01(0.01)	−0.01(0.01)	−0.01(0.01)	0.05(0.04)	0.05(0.04)	0.05(0.04)	−0.06(0.04)	−0.06(0.04)	−0.06(0.04)
DFF	−0.00(0.01)	−0.00(0.01)	−0.00(0.01)	0.04(0.04)	0.04(0.04)	0.04(0.04)	−0.02(0.04)	−0.02(0.04)	−0.02(0.04)
ROA-1	0.10(0.16)	0.10(0.16)	0.10(0.16)	0.38(0.19) *	0.39(0.19) **	0.39(0.19) **	0.36(0.22)	0.36(0.22)	0.36(0.22)
SIZE	−0.00(0.00)	−0.00(0.00)	−0.00(0.00)	0.00(0.01)	0.00(0.01)	0.00(0.01)	0.00(0.01)	0.00(0.01)	0.00(0.01)
LEV	0.02(0.04)	0.02(0.04)	0.02(0.04)	−0.03(0.11)	−0.03(0.11)	−0.03(0.11)	0.11(0.12)	0.11(0.12)	0.11(0.12)
AGE	−0.00(0.00)	−0.00(0.00)	−0.00(0.00)	−0.00(0.00)	−0.00(0.00)	−0.00(0.00)	0.00(0.00)	0.00(0.00)	0.00(0.00)
MKTBK	0.03(0.01) ***	0.03(0.01) ***	0.03(0.01) ***	0.05(0.02) ***	0.05(0.02) ***	0.05(0.02) ***	−0.03(0.02) *	−0.03(0.02) *	−0.03(0.02) *
ENVS	−0.00(0.00)			−0.00(0.00)			−0.00(0.00)
SOCS		−0.00(0.00)			0.00(0.00)			−0.00(0.00)
GOVS			0.00(0.00)			0.00(0.00)			−0.00(0.00)
INDUSTRY
YEAR
obs	206	206	206	205	205	205	206	206	206
R2	0.43	0.43	0.43	0.33	0.33	0.33	0.20	0.20	0.20

***, ** and * indicate that the estimated coefficients are significant at the 1, 5 and 10% levels, respectively.

reports the findings for the base model in Equation (2), estimated using Generalized Least Squares (GLS) with clustering at the firm level and controlling for year and industry fixed effects. Each column shows the results using one of the accrual and real earnings management proxies.

The coefficient for abnormal related party transactions (ARPTs) is positive and significant when negative discretionary accruals is the dependent variable, while Table 4 shows no significant relationship between ARPTs and positive discretionary accruals. With respect to the real earnings management proxies, the results present evidence of a negative relation between ARPTs and abnormal production costs (RMPROD) and a positive association with abnormal discretionary expenses (RMDISX), while they show no relationship between RPTs and abnormal cash flow from operations.

Table 5. Earnings management, related party transactions and ESG scores with interaction effects.

	Model 2		Model 2	Model 2		Model 2	Model 2		Model 2
	NDA		NDA	NDA		PDA	PDA		PDA
INT	−0.08(0.04) *		−0.04(0.03)	−0.05(0.04)		0.28(0.10) ***	0.32(0.10) ***		0.29(0.11) **
ABRPTs	0.17(0.09) *		0.62(0.21) ***	0.42(0.14) ***		0.01(0.33)	−0.53(0.44)		−0.54(0.31) *
BSIZE	−0.00(0.00)		−0.00(0.00)	−0.00(0.00)		0.00(0.00)	0.00(0.00)		0.00(0.00)
WBD	−0.03(0.04)		−0.04(0.04)	−0.03(0.04)		−0.10(0.07)	−0.11(0.07)		−0.09(0.07)
CEOD	−0.02(0.01)		−0.02(0.01)	−0.01(0.01)		−0.01(0.02)	−0.01(0.02)		−0.01(0.02)
DFF	−0.02(0.01) *		−0.02(0.01) **	−0.02(0.01) **		0.01(0.01)	0.01(0.01)		0.02(0.01)
ROA-1	−0.20(0.04) ***		−0.20(0.03) ***	−0.18(0.03) ***		−0.56(0.23) **	−0.56(0.21) **		−0.62(0.23) ***
SIZE	0.01(0.00) **		0.00(0.00) *	0.00(0.00)		−0.01(0.01) *	−0.01(0.01) **		−0.01(0.01) **
LEV	0.01(0.04)		0.02(0.04)	0.02(0.04)		−0.04(0.06)	−0.05(0.06)		−0.04(0.06)
AGE	0.00(0.00)		0.00(0.00)	0.00(0.00)		−0.00(0.00)	−0.00(0.00)		−0.00(0.00)
MKTBK	0.01(0.00) **		0.01(0.00) ***	0.01(0.00) **		0.01(0.01)	0.01(0.01)		0.01(0.01)
ENVS	−0.00(0.00)					−0.00(0.00)
SOCS			−0.00(0.00) ***				0.00(0.00)
GOVS				−0.00(0.00) ***					0.00(0.00)
INDUSTRY
YEAR
RPTs * ENVS	−0.00(0.00)					0.00(0.00)
RPTs * SOCS			−0.01(0.00) ***				0.01(0.01)
RPTs * GOVS				−0.01(0.00) ***					0.01(0.00) *
obs	113		113	113		93	93		93
R2	0.25		0.29	0.29		0.39	0.38		0.38
	Model 2	Model 2	Model 2	Model 2	Model 2	Model 2	Model 2	Model 2	Model 2
	RMCFO	RMCFO	RMCFO	RMPROD	RMPROD	RMPROD	RMDIX	RMDIX	RMDIX
INT	0.01(0.05)	0.04(0.05)	0.03(0.06)	−0.14(0.16)	−0.13(0.17)	−0.14(0.17)	0.05(0.19)	0.05(0.19)	0.05(0.19)
ABRPTs	0.38(0.12) ***	0.43(0.31)	0.18(0.19)	−0.13(0.22)	−0.30(0.61)	−0.47(0.34)	0.26(0.24)	0.41(0.63)	0.47(0.35)
BSIZE	−0.00(0.00)	−0.00(0.00)	−0.00(0.00)	0.00(0.00)	0.00(0.00)	0.00(0.00)	−0.00(0.00)	−0.00(0.00)	−0.00(0.00)
WBD	−0.07(0.04) *	−0.08(0.04) *	−0.08(0.04) *	−0.06(0.15)	−0.06(0.15)	−0.06(0.15)	−0.02(0.15)	−0.02(0.15)	−0.02(0.15)
CEOD	−0.02(0.01)	−0.01(0.01)	−0.01(0.01)	0.05(0.04)	0.05(0.04)	0.05(0.04)	−0.06(0.04)	−0.06(0.04)	−0.06(0.04)
DFF	−0.01(0.01)	−0.00(0.01)	−0.00(0.01)	0.04(0.04)	0.04(0.04)	0.04(0.04)	−0.02(0.04)	−0.02(0.04)	−0.02(0.04)
ROA-1	0.04(0.13)	0.08(0.15)	0.10(0.15)	0.36(0.19) *	0.39(0.19) **	0.39(0.19) **	0.37(0.22) *	0.35(0.22)	0.04(0.22)
SIZE	0.00(0.00)	−0.00(0.00)	−0.00(0.00)	0.00(0.01)	0.00(0.01)	0.00(0.01)	0.00(0.01)	0.00(0.01)	0.00(0.01)
LEV	0.00(0.04)	0.02(0.04)	0.02(0.04)	−0.03(0.12)	−0.03(0.11)	−0.04(0.12)	0.11(0.12)	0.11(0.12)	0.11(0.12)
AGE	−0.00(0.00)	−0.00(0.00)	−0.00(0.00)	−0.00(0.00)	−0.00(0.00)	−0.00(0.00)	0.00(0.00)	0.00(0.00)	0.00(0.00)
MKTBK	0.03(0.01) ***	0.03(0.01) ***	0.03(0.01) ***	0.05(0.02) ***	0.05(0.02) ***	0.05(0.02) **	−0.03(0.02) *	−0.03(0.02) *	−0.03(0.02) *
ENVS	−0.00(0.00) ***			−0.00(0.00)			0.00(0.00)
SOCS		−0.00(0.00)			0.00(0.00)			−0.00(0.00)
GOVS			−0.00(0.00)			0.00(0.00)			−0.00(0.00)
INDUSTRY
YEAR
RPTS * ENVS	−0.01(0.00) ***			−0.00(0.00)			0.00(0.00)
RPTS * SOCS		−0.01(0.00)			0.00(0.01)			−0.00(0.01)
RPTS * GOVS			−0.00(0.00)			0.00(0.01)			−0.00(0.01)
obs	206	206	206	205	205	205	206	206	206
R2	0.47	0.44	0.43	0.33	0.33	0.33	0.20	0.20	0.20

***, ** and * indicate that the estimated coefficients are significant at the 1, 5 and 10% levels, respectively.

reports the findings for the model in Equation (3), taking into account the interaction effects of RPTs and Environmental, Social and Governance scores. The results confirm that negative discretionary accruals are positively associated with abnormal related party transactions; they also show a negative interaction effect between ARPTs and ESG scores. For what concerns real earnings management, the inclusion of the interaction effects produces only a weak relation and interaction with RPTs for abnormal cash flow from operations. This model, similarly as the previous one, includes dummy variables to control for differential effects related to industrial peculiarities and time series variability.

Table 6. Hypotheses testing.

Hypotheses	Dependent Variable	Independent Variable	Expected. Sign	Value	t	Decision
H1a. Abnormal relates party transactions are associated to real earnings management	RMCFO	ABRPTs	+/−	−0.07	0.09	Supported for REM via production cost and discretionary expenses
	RMPROD	ABRPTs	+/−	−0.27	0.10 ***
	RMDIX	ABRPTs	+/−	+0.31	0.11 ***
H1b. Abnormal relates party transactions are associated to accrual earnings management	NDA	ABRPTs	+/−	+0.06	0.03 **	Supported for downward AEM
	PDA	ABRPTs	+/−	−0.02	0.10
H2a. ESG performance moderates the relation between relates party transactions and real earnings management	RMCFO	RPTs * ENVS	+/−	−0.01	0.00 ***	Supported for Environmental performance and REM via cash flow
		RPTs * SOCS	+/−	−0.01	0.00
		RPTs * GOVS	+/−	−0.00	0.00
	RMPROD	RPTs * ENVS	+/−	−0.00	0.00
		RPTs * SOCS	+/−	+0.00	0.01
		RPTs * GOVS	+/−	+0.00	0.01
	RMDIX	RPTs * ENVS	+/−	+0.00	0.00
		RPTs * SOCS	+/−	−0.00	0.01
		RPTs * GOVS	+/−	−0.00	0.01
H2b. ESG performance moderates the relation between relates party transactions and accrual earnings management	NDA	RPTs * ENVS	+/−	−0.00	0.00	Supported for Social and Governance performance and downward AEM Supported for Governance performance and upward AEM
		RPTs * SOCS	+/−	−0.01	0.00 ***
		RPTs * GOVS	+/−	−0.01	0.00 ***
	PDA	RPTs * ENVS	+/−	+0.00	0.00
		RPTs * SOCS	+/−	+0.01	0.01
		RPTs * GOVS	+/−	+0.01	0.00 *

***, ** and * indicate that the estimated coefficients are significant at the 1, 5 and 10% levels, respectively.

displays the results of hypotheses testing.

4.3. Discussion

The results highlight different relations between RPTs and real earnings management according to the type of management. They indicate a not significant positive relation between RPTs and RMCFO. The association becomes significant only controlling for the effect of environmental performance. This reveals that RPTs are not significantly related to REM practices perpetrated by anticipating sales through increased price discounts and/or by granting more advantageous credit conditions to customers, suggesting that abnormal levels of RPTs might be used also as an autonomous form of earnings manipulation.

The findings show a negative significant relation between RPTs and RMPROD. This result suggests that firms with high levels of RPTs are less prone to boost earnings by the means of overproduction and the consequent reduction of fixed cost per unit sold. In fact, real earnings management practices are costly for shareholders [90] as companies manipulating their operations to increase earnings might divert from normal business practices and face a decline in their subsequent operating performance [91]. Although managers might meet short-term earnings targets by the means of REM [92], they might put at risk a firm’s long run value. In cases of overproduction, a firm will incur greater inventory holding costs and lower liquidity [4]. This may explain the empirical evidence suggesting that companies use RPTs as a substitute for production cost-based earnings management.

Conversely, the results show a significant positive association between RPTs and RMDIX, suggesting that firms engaging in abnormal levels of transactions with related parties tend, at the same time, to increase earnings, lowering discretionary expenses. When managers use RPTs in order to expropriate a company’s resources, they have incentives to engage in earnings management to increase the benefits extracted or to mask such expropriation [2,93]. Therefore, in this case, they might be motivated to reduce discretionary expenses in order to increase current period earnings, as the possible negative effect of this manipulation is likely to occur over a longer period with respect to an increase in costs from overproduction. H1 is completely verified in cases of real earnings management perpetrated through the manipulation of the production cost and of discretionary expenses.

The results point out a significant positive relation between RPTs and downward accrual-based earnings management. This evidence suggests that firms engaging in RPTs to transfer benefits to the controlling shareholders also use accrual-based earnings management to reduce dividends, raise self-financing and avoid the use of external equity financing [29].

The findings indicate a not significant negative association between RPTs and upward accrual earnings management. The relationship becomes significant only when controlling for the effect of governance performance. This suggests that firms that have high levels of RPTs do not significantly use these operations as a substitute for upward earnings management, although accrual-based earnings manipulation is harder to detect [55]. Therefore, H2 is completely verified only in cases of downward accrual-based earnings management when earnings manipulation is related to second type agency conflicts.

Previous research did not find a significant relationship [21] or a significant negative relation [22] between RPTs and AEM. These studies measured AEM without distinguishing downward and upward accrual earnings management, and this might explain El-Helaly et al. [21] results, as the lack of a significant relationship might to be due to an offsetting effect. Consistently, the result of Alhadab et al. [22] is expected to be due to the peculiar characteristics of the Jordan business environment [22] that may result, overall, in a negative relation between RPTs and AEM.

The results do not point out a significant moderating effect of ESG performance on the relation between RPTs and real earnings management. They only suggest that environmental performance moderates the possible use of RPTs in association with earnings manipulations through sales anticipations. Therefore, the findings support H2a only in the case of cash flow earnings manipulations.

Conversely, the analysis of the impact of ESG performance points out an overall moderating effect on the relation between RPTs and AEM, supporting H2b. More specifically, social and governance performance exerts a moderating effect on the association between RPTs and downward earnings management. This suggests that more socially responsible firms are less prone to use RPTs as an additional form of earnings management or a method to mask the expropriation of minority shareholders to the benefit of controlling stockholders. This effect is consistent with the empirical evidence that firms more committed to social activities are less prone to manipulate their financial information [13,37]. Social aspects are an external factor that motivate firms to act in order to preserve their image [94], and socially oriented companies prefer to develop long-term relationships with their stakeholders whilst AEM practices conflict with this attitude as they harm shareholders’ interests [64]. Moreover, ESG performance increases stakeholders’ expectations of a firm’s behavior, and AEM might undermine stakeholders’ trust and produce a pragmatic and moral legitimacy gap, jeopardizing a firm’s durability [42,68].

The same moderating effect holds for firms with good systems and processes to ensure that their management behaves in the interest of their shareholders at large. The results also indicate that good governance mechanisms moderate the possible use of RPTs as a substitute for upward accrual-based earnings management. This is consistent with empirical evidence pointing out that governance performance strongly moderates the use of AEM [16].

These findings are consistent with the view that the concern for ESG is a form of management responsibility that can reduce agency costs [68] stemming from the association between RPTs and accrual earnings management.

5. Conclusions

This study investigates the relation between RPTs and different types of earnings management as well as the possible moderating effect of ESG performance, analyzing a sample of Italian non-financial listed firms for the period 2014–2019.

It contributes to the debate on RPTs and agency conflicts, providing evidence that firms might use RPTs in association with downward AEM, as a substitute for REM perpetrated by sales anticipations and discretionary expenses reductions, as well as an autonomous form of earnings manipulation. This research provides supporting information for the Stakeholder theory and legitimacy perspective as the empirical evidence shows a significant moderating effect of ESG performance on earnings alterations. In particular, social as well as governance performance significantly moderates the association between RPTs and downward AEM, whilst environmental performance moderates the possible use of RPTs in association with cash flow-based REM. These results have theoretical implications as they highlight, for the first time, the moderating role of ESG performance on how agency conflicts shape the association between RPTs and EM.

These results have implications for regulators as they suggest the opportunity to strengthen policies to boost companies’ commitment in social and environmental activities given the moderating effect on the association between the aforementioned forms of earnings manipulations. These findings have implications for market participants as ESG performance is an indicator of a more ethical corporate behavior and financial information reliability. This implies that companies might enhance investor trust in their financial reporting practices and, more generally, in their decision-making process by means of their ESG performance.

This explorative study presents some limitations. It analyzes a single country sample in order to focus on the effect of Environmental, Social and Governance performance on the association between RPTs and EM, avoiding possible interferences related to different institutional settings. Hence, the results may not be generalizable. Recently, the literature has pointed out differences in earnings management practices across countries [95,96]; therefore, further research could focus on an international sample in order to analyze how specific institutional characteristics may influence ESG performance’s moderating effect.

This research used ownership type as a control variable, but it did not analyze how this characteristic might affect the moderating role of ESG performance on the relation between RPTs and EM; therefore, future studies could investigate this aspect.

Moreover, it focused on the three ESG pillars, but future research might develop more grained analyses disentangling the effect of specific sub-pillars.