The impact of ESG performance on earnings quality: evidence from Italy

During recent decades, the debate on earnings management (EM) has gained prominence among academics, practitioners and regulators. Major financial scandals have intensified scrutiny over managerial opportunism and earnings manipulation (Md Nasir et al., 2018). These issues can affect the entire economic system, jeopardising trust in financial markets and institutions (Lotfi et al., 2022); they have raised concerns about earnings quality (EQ), defined as the extent to which reported income reflects companies' true financial performance (Dechow et al., 2010). In this study, the terms earnings management and earnings manipulation are used interchangeably to refer to opportunistic financial reporting behaviour.

Recent regulatory changes and stakeholder pressure have increased attention on the role of ESG practices in financial reporting reliability.

Prior evidence is mixed, suggesting that ESG may either constrain (Kim et al., 2012; Tohang et al., 2024) or, conversely, obscure earnings manipulation (Dhaliwal et al., 2011; García-Sánchez et al., 2020), highlighting the need for context-specific analysis.

Against this background, this study explores the relationship between ESG performance and EQ in large Italian listed companies, assessing whether firms with stronger ESG commitment exhibit higher EQ.

The research advances the literature in several ways. First, while prior research has largely examined corporate social responsibility (CSR) disclosure or discretionary accruals (Velte, 2019), we rely on quantitative ESG performance metrics and employ the Beneish M-Score, which captures both accrual-based and structural manipulation. Second, we disaggregate ESG into its environmental, social and governance components to identify the dominant drivers of EQ. Moreover, most existing evidence is based on Anglo-Saxon markets, which differ substantially from the stakeholder-centric institutional context typical of Italy, which is also characterised by high ownership concentration, often family-based (La Porta et al., 2000).

The remainder of the paper is structured as follows: Section 2 presents the theoretical background and literature review; Section 3 describes the methodology and variables we used; Section 4 reports the empirical results; Section 5 discusses the findings, while Section 6 outlines implications, limitations and directions for future research.

Corporate non-financial responsibility has evolved from CSR's ethical focus to sustainability integration and reputational concerns rather than measurable impacts (Carroll, 1999). In the late 1990s, the concept of corporate sustainability focused on the integration of environmental and social factors into corporate strategy and business models, often summarised by the triple bottom line (Elkington, 1997). During the 2000s, the growing demand from investors and regulators for comparable and auditable non-financial information pushed companies toward the transition to the ESG paradigm (Eccles and Klimenko, 2019). ESG data are increasingly used to evaluate companies' exposure to non-financial risks and long-term resilience (Eccles et al., 2020).

In general, ESG operationalises CSR and sustainability into standardised, investor-relevant metrics, scores, ratings and reporting principles (Kotsantonis et al., 2016). The main differences between corporate social responsibility, sustainability and ESG factors are summarised in Table 1.

Given our empirical design, ESG scores provide consistent, time-varying and comparable measures suitable for panel analysis.

Several complementary theoretical frameworks help explain how ESG performance influences EQ and reporting reliability. Agency theory (Jensen and Meckling, 1976) suggests that managers might “cook the books” to serve their interests and personal gain to the detriment of shareholders.

Other studies exhibit a positive relationship between ESG compliance and EM, since managers can undertake sustainability activities and ESG actions to obtain personal advantages (Dhaliwal et al., 2011). They might report ESG and sustainability performance to conceal opportunistic behaviours and EM practices (Velte, 2024). As ESG disclosure can be a tool to enhance manager reputation and stakeholder trust, it could also enable companies to manage their earnings and provide unreliable, low-quality financial statements (Kim et al., 2012).

Stakeholder theory (Freeman, 1984), on the other hand, expands the fiduciary responsibility beyond shareholders, arguing that businesses must consider the needs and interests of all stakeholders. Firms with strong ESG practices often exhibit higher levels of engagement and accountability toward stakeholders, which can in turn enhance financial reporting quality and reduce EM (Mutuc et al., 2019).

Legitimacy theory (Suchman, 1995) posits that firms undertake socially responsible actions to maintain and strengthen their legitimacy within society. High-quality financial reporting can be viewed as legitimacy-seeking behaviour, and ESG activities may reinforce this perception. Since EM can result in a threat to business from society, it represents a risk to the company's legitimacy and to its survival (Grougiou et al., 2014). In this perspective, scholars also assume a negative relationship between ESG reporting and earnings manipulation, as a higher degree of disclosure can diminish managerial incentives to engage in EM practices (Vatis et al., 2025).

More recent contributions have further consolidated the correlation between ESG factors and EQ. A comprehensive meta-systematic review by Lim et al. (2022) has shown how ESG and the total quality management approach can be integrated. These two elements, in fact, share convergent principles of transparency, stakeholder engagement and continuous improvement, reinforcing the notion that ESG-oriented cultures promote higher reporting integrity and performance consistency.

EM has been extensively analysed due to its implications for financial reporting reliability and stakeholder decision-making. EQ reflects the extent to which reported earnings capture companies' underlying performance and future prospects (Dechow and Dichev, 2002), and is influenced by information asymmetry and managerial discretion.

Despite the growing interest in the relation between EQ, CSR, sustainability and ESG compliance within institutions, practitioners and academics, empirical evidence remains mixed.

Velte (2019) points out a negative relationship between ESG scores and EM, implying that businesses with stronger ESG performance are less likely to undertake manipulative accounting actions. Similarly, Lestari and Muthmainnah (2025) show that ESG performance can mitigate informational distortions and moderates the relationship between EM and market reaction to corporate earnings information. Companies committed to sustainable policies adopt a more conservative approach regarding their financial statements, as managers have a long-term perspective regarding profits and, consequently, they try to avoid actions that could negatively affect their reputation among investors and stakeholders.

Sectoral and country-specific analyses confirm the multidimensional nature of the relationship between ESG variables and EQ. Focusing on energy and utilities firms, Persakis et al. (2025) show that ESG compliance is positively linked to EQ and financial reporting reliability. Analysing Chinese listed firms, Liu et al. (2025) reveal that the long-term orientation embedded in high ESG performance reduces both accrual-based and real EM.

From another perspective, some pieces of research underscore that managers only engage in sustainability and ESG compliance activities if they can achieve personal benefits (Dhaliwal et al., 2011). Managers may adopt ESG practices to gain media coverage and societal legitimacy while concealing their true activities from shareholders and carrying out “socialwashing” or “greenwashing” practices (García-Sánchez et al., 2020).

Mutuc et al. (2020) point out that sustainability accomplishments may push EM because of stakeholder pressure for better performance. Firms pursuing high sustainability performance are inclined to face large investments, which may decrease earnings and the company's ability to reach its goal in terms of profitability. Examining French and Spanish listed companies, Borralho et al. (2022) confirm that ESG commitment has been exploited to deceive stakeholders and divert their attention from financial reporting content.

To sharpen the focus on the relationship between ESG performance and EQ, Table 2 summarises the main empirical studies that have examined this link in different institutional settings.

Overall, prior findings remain context-dependent, reinforcing the relevance of examining the Italian setting, where large, listed firms operate under the EU's evolving sustainability reporting regime but still face ownership concentration and informational opacity that may influence financial reporting quality.

Literature has also offered insights into the influence of single ESG factors on EQ. According to Adeneye et al. (2024), there is a positive correlation between EQ, environmental and social pillars, supporting previous research affirming that managers who promote socially responsible actions are less prone to indulging in EM practices (Ehsan et al., 2022). Focusing on governance factors such as board characteristics, Persakis et al. (2025) assert that they strengthen the positive impact of ESG performance on financial reporting quality. Among ESG sub-dimensions, Liu et al. (2025) also state that the social and governance “pillars” have the strongest influence on EM, whereas the environmental variable is less influential. The effects of the single ESG components on EQ based on different theoretical frameworks have been summarised in Figure 1.

In line with Löw and Cordovez (2023), we adopt ESG performance also as a proxy for corporate sustainability and CSR engagement, recognising that although conceptually distinct, these constructs are often empirically interchangeable. We interpret ESG performance as a broader measurement than sustainability, which assesses how a business acts in a sustainable and responsible manner, relying on three main components (Gillan et al., 2021). Notwithstanding the mixed results achieved by previous literature, drawing primarily on stakeholder and legitimacy theory, we argue that firms committed to ESG practices are less prone to engage in opportunistic EM, enhancing EQ, financial reporting reliability and transparency (Velte, 2019; Nurrahman et al., 2019).

Accordingly, we propose:

Emerging literature suggests that the three ESG dimensions may exert different effects on financial reporting practices (Block and Wagner, 2014). For instance, strong governance may directly reduce managerial discretion, while environmental and social efforts may indirectly foster stakeholder trust and financial reporting reliability.

Our second hypothesis can thus be expressed as follows:

To test our hypotheses, we use a panel data regression model utilising a sample of 51 companies over a three-year period (2021–2023), resulting in a balanced panel of 153 firm-year observations.

The adoption of a short panel (2021–2023) is motivated by the relevance of this timeframe, which captures a critical phase characterised by post-COVID-19 recovery and by the progressive implementation of European sustainability regulations, including the transition towards the Corporate Sustainability Reporting Directive (CSRD). During this period, ESG practices have become increasingly embedded in corporate governance and reporting systems, making it particularly suitable for examining the relationship between ESG performance and EQ.

A potential concern in this analysis is reverse causality, whereby firms with inherently higher EQ may also exhibit stronger ESG performance. To mitigate this issue, we employ fixed-effects estimators (Wooldridge, 2010) and also lagged ESG variables.

A fixed-effects panel regression model is employed to control for unobservable, time-invariant firm-specific characteristics – such as ownership structure, managerial culture and governance traditions – that may influence EQ. Fixed-effects estimators are widely employed in accounting and corporate governance research when firm-level heterogeneity is expected to be correlated with explanatory variables (Baltagi, 2021). The choice of fixed effects over alternative estimators, such as pooled ordinary least squares (OLS) or random effects, is supported by the Hausman test (Hausman, 1978), which indicates that firm-specific effects are correlated with the regressors, making the fixed-effects specification the most appropriate for this analysis.

EQ is proxied by the Beneish M-Score (1997,1999). Although this tool does not measure EQ directly, it provides a comprehensive indicator of manipulation risks by capturing both accrual-based and real EM practices. Recent empirical studies have increasingly adopted this approach to examine the relationship between ESG-related factors and financial reporting quality (Borralho et al., 2022; Persakis et al., 2025). A score below −2.22 indicates a low probability of manipulation, while higher values signal greater manipulation risk. Compared to discretionary accrual models, this approach offers a broader assessment of financial reporting behaviour, especially in contexts characterised by complex ownership structures and potential opacity (Dechow et al., 2010).

Regarding our first hypothesis, the main test variable is firm performance in sustainability and ESG fields, proxied by LSEG ESG scores (Dorfleitner et al., 2015), which provide standardised, time-varying and cross-firm comparable indicators of ESG performance (LSEG, 2024). These scores are widely used in empirical ESG research due to their methodological transparency, sector-specific weighting schemes and suitability for panel data analysis (Kotsantonis et al., 2016; Velte, 2019). The database also provides the score for each of the three pillars we employed to test our second hypothesis.

Given the structure of the Beneish M-Score, we expect ESG variables to exhibit negative coefficients, indicating higher EQ.

We also included size, leverage, profitability and growth as control variables.

Following previous research (Rezaee and Tuo, 2019), firm size has been defined by the natural logarithm of total assets. Researchers have not reached a unanimous consensus on the correlation between companies' size and EQ. Recent works, however, highlight a positive correlation between firm size and EQ due to stronger monitoring and disclosure requirements compared to smaller companies (Ashbaugh-Skaife et al., 2007; Ben Amar and Chakroun, 2018).

In our research, correspondingly, we hypothesise a positive relationship between size and EQ.

Financial leverage has been determined by the total liabilities/total assets ratio. Highly leveraged firms may face pressure to meet debt covenants, potentially increasing earnings manipulation (Dechow et al., 1996; Gong et al., 2008).

Accordingly, we assume a negative relationship between leverage and EQ.

Profitability, proxied by the return on sales (ROS) has been considered negatively correlated with EM (Doyle et al., 2007). Lower profitability, in fact, can push management to overestimate revenues and understate expenses (Kreutzfeldt and Wallace, 1986). Companies with higher margins, on the other hand, do not have strong incentives to manipulate their earnings (Dimitropoulos, 2022).

Consequently, we expect a positive correlation between profitability and EQ.

In the light of previous studies (Beasley, 1996), we also considered firm growth, which has been proxied by Tobin's Q (1978), following Fassas et al. (2023). Consistent with Bell et al. (1991), excessively high growth can be considered a “red flag”, indicating a higher likelihood of earnings manipulation, since managers may be driven to forge financial statements to sustain market expectations and communicate stability and low volatility.

Therefore, we expect a negative relationship between growth and EQ.

In the light of the above-mentioned considerations, the model related to our first hypothesis can be synthesised as follows:

Our second hypothesis, on the other hand, has been tested relying on the following regression model:

Due to the nature of our dependent variable, we summarise the expected sign of independent variables' coefficients in Table 3. It is worth observing that a negative coefficient reveals higher EQ because the Beneish M-Score decreases and vice versa.

This section presents the empirical findings of the study. We first provide descriptive statistics to illustrate the main characteristics of the sample and the distribution of the variables. We then discuss the results of the regression results for the overall ESG score (Model 1) and the individual ESG pillars (Model 2).

Table 4 reports the descriptive statistics for the main variables included in the analysis. The Beneish M-Score exhibits considerable dispersion, indicating substantial heterogeneity in earnings manipulation risk across firms and years. ESG scores also display a wide range, both for the aggregate indicator and for the individual ESG pillars. Among the ESG dimensions, the environmental score shows the highest standard deviation, indicating greater variability in firms' environmental performance within the sample.

Table 5 summarises the results of Model 1, which tests the relationship between overall ESG performance and EQ.

ESG score (β = −0.115, p < 0.05) is statistically significant and its negative coefficient aligns with expectations, since lower Beneish values indicate higher EQ. More specifically, this result shows that a one-point increase in the ESG score is associated with a 0.115 decrease in the Beneish M-Score, reducing the likelihood of earnings manipulation. This outcome supports H1, according to which companies more committed to sustainability and ESG compliance show a higher level of EQ and are less likely to engage in earnings manipulation practices.

Regarding control variables, only size (β = 1.152, p < 0.05) is statistically significant, although it does not exhibit the sign we expected, indicating that larger firms are associated with higher Beneish M-Scores and, therefore, a greater likelihood of earnings manipulation.

Profitability (β = −0.050, p > 0.10) and growth (β = 0.597, p > 0.10) show the expected coefficient sign but are not statistically significant. Finally, the leverage coefficient sign (β = −2.506, p > 0.10) is not in line with our expectations and it is not statistically significant, revealing that this variable does not affect EM activities.

Overall, the results of model 1 denote that ESG performance plays a significant role in explaining the variation in EQ, in line with our expectations, whereas most control variables do not exert a statistically significant effect.

Table 6 presents the results of Model 2, which disaggregates ESG performance into its environmental (ENV), social (SOC) and governance (GOV) components.

The environmental score (ENV) displays a negative coefficient (β = −0.035) but is not statistically significant (p > 0.10), pointing out that environmental performance does not exert a measurable influence on earnings manipulation within the sample. By contrast, as we expected, both the social (SOC) and governance (GOV) components are negative and statistically significant, with SOC (β = −0.136, p < 0.05) exerting the largest effect among the three dimensions, followed by GOV (β = −0.086, p < 0.05).

These results indicate that stronger social engagement and more robust governance mechanisms are primary drivers of the ESG–EQ relationship.

Control variables behave similarly to model 1: size retains its positive and marginally significant coefficient (β = 1.113, p < 0.10), signalling that larger firms remain more prone to earnings manipulation also when considering the individual ESG pillars.

The other statistically significant test variable is growth (β = 2.497, p < 0.05), which exhibits the coefficient sign we assumed (with a much higher value in model 2), demonstrating that high growth companies are more likely to indulge in EM practices than mature or more stable firms, consistently with previous research (Bell et al., 1991).

Profitability (β = 0.071, p > 0.10) and leverage (β = 1.787, p > 0.10) again show no statistically significant effects.

Overall, Model 2 confirms that the ESG–EQ relationship is mainly driven by the social and governance dimensions, rather than by environmental performance.

Model 1 displays an R² of 0.423 and an adjusted R² of 0.324 (Table 5), indicating that roughly one-third of the variation in the Beneish M-Score is explained by ESG performance and the control variables. R² values are consistent with prior accounting and ESG research, where non-financial constructs typically explain a limited portion of EQ variation (Martínez-Ferrero and García-Sánchez, 2018; Velte, 2019). The standard error of the estimate (SEE = 1.480) confirms an adequate level of precision and is consistent with empirical work using Beneish-type specifications (Beneish, 1997, 1999).

Model 2 shows slightly lower explanatory power, with an R² of 0.348 and an adjusted R² of 0.210 (Table 6), as expected when disaggregating ESG into its individual pillars. This reduction aligns with the observation that single factors tend to be weaker predictors of earnings manipulation than the aggregate ESG construct (Block and Wagner, 2014; Borralho et al., 2022). This level of explanatory power is reasonable given the multidimensional nature of ESG performance and the inherent complexity of EM practices.

Variance inflation factor (VIF) values range between 1 and 5, indicating no serious multicollinearity concerns (Allison, 1999). Overall, diagnostic checks confirm the adequacy of the regression specifications.

To further assess the robustness of the baseline findings and to address potential endogeneity concerns, additional model specifications were considered. In particular, we re-estimated the regression models using one-year lagged ESG variables (both overall score and individual pillar values). This approach is commonly adopted in panel data settings to mitigate simultaneity and reverse causality issues by introducing a temporal separation between the explanatory and dependent variables (Wooldridge, 2010; Baltagi, 2021).

The results indicate a reduction in explanatory power, as reflected in a lower R² and a decrease in the statistical significance of the ESG coefficients compared to the baseline model. While these findings weaken the predictive strength of ESG performance in a lagged setting, they provide additional insight into the nature of the relationship. Therefore, the evidence suggests that ESG performance is more strongly associated with contemporaneous EQ, rather than exerting a delayed effect. This interpretation is consistent with the view that ESG practices operate as ongoing governance and monitoring mechanisms, influencing financial reporting behaviour in real time rather than through persistent intertemporal effects (Dechow et al., 2010; Velte, 2019).

The reduction in statistical significance in the lagged specification may also reflect the relatively short time horizon of the analysis (2021–2023), which limits the ability to capture longer-term ESG effects.

This study examines the relationship between ESG performance and EQ in large Italian listed companies, contributing to the literature on whether sustainability practices enhance financial reporting reliability. The results show that stronger ESG performance is associated with higher EQ, proxied by a lower Beneish M-Score.

From a theoretical perspective, this relationship can be jointly interpreted through stakeholder and legitimacy theory, which offer complementary explanations of the same underlying mechanism. Stakeholder theory suggests that firms accountable to a broad set of stakeholders face greater reputational and economic constraints, discouraging opportunistic reporting (Freeman, 1984). Legitimacy theory highlights that firms seek social acceptance by aligning with societal expectations (Suchman, 1995), with transparent reporting serving as a key legitimacy mechanism. In this context, transparent and reliable financial reporting serves as a legitimacy-enhancing mechanism (Grougiou et al., 2014). Taken together, these frameworks suggest that ESG engagement increases the costs of EM by amplifying stakeholder scrutiny and legitimacy risks, thereby fostering higher EQ.

These findings are consistent with prior international evidence showing that ESG-oriented firms tend to adopt more transparent and reliable reporting practices (Velte, 2019; Tohang et al., 2024). ESG compliance appears to strengthen monitoring mechanisms, reduce information asymmetry and enhance managerial accountability, ultimately constraining discretionary reporting behaviour. ESG performance can therefore operate as a governance-enhancing mechanism that complements formal control systems.

The disaggregated analysis in Model 2 provides further insight into which ESG pillar most strongly influences EQ. While the environmental dimension is not statistically significant, both social and governance components show a strong negative association with earnings manipulation. The social pillar emerges as the most influential dimension, followed by governance, indicating that stakeholder-oriented practices and formal governance mechanisms play a central role in shaping financial reporting behaviour.

The prominence of the social dimension is consistent with prior studies suggesting that firms investing in employee relations, human capital management and stakeholder engagement face higher reputational and legitimacy costs in the event of financial misreporting (Mutuc et al., 2019; Ehsan et al., 2022). Strong social performance intensifies firms' exposure to stakeholder scrutiny and relational accountability, making opportunistic EM more costly and less sustainable over time. This result aligns with evidence showing that socially responsible firms are less prone to accrual-based and real earnings manipulation (Martínez-Ferrero et al., 2015; Adeneye et al., 2024).

The governance component also exhibits a significant negative association with earnings manipulation, confirming its fundamental role in mitigating agency problems. Governance mechanisms – such as board effectiveness, internal controls and transparency practices – directly constrain managerial discretion and reduce opportunities for EM (Jensen and Meckling, 1976; Gillan et al., 2021). This finding is consistent with empirical evidence indicating that governance is often the ESG dimension most closely linked to financial reporting quality (Velte, 2019; Rahman et al., 2024; Persakis et al., 2025). From an agency-theoretic standpoint, governance represents the most immediate and formal mechanism through which ESG engagement translates into improved EQ.

By contrast, the lack of statistical significance for the environmental component suggests that environmental performance does not exert a measurable influence on EM within the sample. This outcome may reflect the limited short-term financial materiality of environmental initiatives or their weaker integration into accounting and reporting systems. Environmental investments, in fact, typically generate long-term benefits and are less directly linked to short-term financial reporting practices. As a result, their impact on earnings manipulation may not be immediately observable within a relatively short time span (Eccles and Klimenko, 2019; Velte, 2019). From a stakeholder–legitimacy perspective, environmental performance may generate legitimacy benefits that are more external-facing and long-term in nature, with less direct implications for financial reporting credibility. From an institutional perspective, in the Italian context, investors and regulators may place stronger emphasis on governance strength and social accountability – such as transparency, compliance and employee relations – than on environmental performance when assessing reporting reliability. Consequently, environmental initiatives may contribute more to external legitimacy and long-term reputation than to the immediate discipline of financial reporting practices (Kotsantonis et al., 2016; Eccles et al., 2020). This interpretation is consistent with prior literature suggesting that the social and governance dimensions are more strictly linked to mechanisms that directly constrain managerial discretion and earnings manipulation (Dechow et al., 2010; Velte, 2019).

The institutional setting of Italy provides an important lens through which these findings can be interpreted. Italy is characterised by a stakeholder-oriented governance model, high ownership concentration and a regulatory environment shaped by mandatory non-financial disclosure under EU directives. In such a context, social engagement and governance quality may be more salient to stakeholders and regulators than environmental initiatives, thereby exerting a stronger disciplinary effect on managerial reporting behaviour.

Regarding control variables, the positive association between firm size and earnings manipulation contrasts with the conventional expectation that larger firms, being more visible and subject to greater scrutiny, should exhibit higher reporting quality (Francis et al., 2008). In the Italian setting, large firms are often characterised by concentrated ownership, family control and complex organisational structures, which may weaken effective monitoring despite formal governance mechanisms (La Porta et al., 2000). Greater organisational complexity and multiple managerial layers may increase information asymmetry and provide additional scope for discretionary accounting choices, offsetting the potential monitoring benefits associated with firm size (Gaio and Raposo, 2011).

Overall, the results indicate that ESG performance enhances EQ primarily through mechanisms related to stakeholder accountability and governance effectiveness, rather than through environmental initiatives alone. These findings contribute to the ongoing debate on whether ESG practices represent substantive governance tools or symbolic compliance strategies, providing evidence in favour of the former – particularly with respect to the social and governance dimensions. By highlighting the differential impact of ESG pillars, this study underscores the importance of moving beyond aggregate ESG scores to better understand how specific ESG components influence financial reporting reliability.

This study examines the relationship between ESG performance and EQ in large Italian listed companies over the period 2021–2023. Using a fixed-effects panel regression model and the Beneish M-Score as a proxy for earnings manipulation, the analysis provides evidence of a significant positive association between ESG performance and EQ.

From a theoretical perspective, our findings contribute to the literature linking ESG performance and financial reporting quality, supporting agency, stakeholder and legitimacy theories. The empirical results suggest that firms committed to ESG compliance are more likely to align their behaviour with broader societal expectations, thus reducing information asymmetry and enhancing financial statements' trustworthiness.

From a quality-management perspective, ESG practices function as organisational control systems that enhance reporting reliability, reduce process opportunism and support continuous improvement in financial disclosure.

From a managerial viewpoint, the findings signal that ESG integration represents more than just a compliance-oriented or reputational strategy. Firms with stronger ESG performance – particularly in the social and governance dimensions – exhibit higher EQ, indicating that ESG engagement can strengthen internal controls and improve financial reporting reliability. Governance arises as a tangible variable through which ESG engagement translates into improved EQ, confirming that institutional monitoring, corporate governance mechanisms and stakeholder accountability reduce managerial discretion.

Managers and financial officers should therefore view ESG practices as effective governance-enhancing tools that can reduce information asymmetry, strengthen credibility with investors and support long-term value creation.

Our findings provide important insights for policymakers and regulators, since their actions aimed at pushing firms to pay greater attention to sustainability and socially responsible actions could, in turn, also reduce information asymmetry, lower earnings manipulation and increase overall market trust.

The results, therefore, also encourage policymakers and standard setters to promote more consistent, financially relevant environmental metrics, bridging the current gap between sustainability communication, reporting comparability and accounting transparency.

One of the main limitations of the study concerns the sampling procedure, since we only examined large, listed companies that are also subject to the legal obligation of issuing a sustainability report, which often incentivises improved sustainability and ESG performance.

Moreover, the relatively short time horizon may constrain the ability to capture long-term ESG effects and limits the strength of causal inference. Regarding this issue, although lagged ESG specifications were considered to address potential endogeneity concerns, the associated reduction in explanatory power and statistical significance suggests that the temporal dynamics of the ESG–EQ relationship require further investigation.

Future research should extend the analysis over longer periods and employ more advanced econometric techniques to more effectively resolve causality problems and strengthen the robustness of the outcomes.

From a methodological point of view, alternative proxies for EQ such as discretionary accruals or earnings persistence, could be employed in future research to further validate the robustness of the findings.

As for the variables taken into consideration, while the model includes the main firm-level determinants commonly adopted in the literature, additional controls such as audit quality could further refine the specification. Due to data constraints, this variable is not included in the current analysis and is left for future research. The impact of ESG commitment on EQ can also be affected by sector-specific dynamics. For example, heavily regulated sectors such as energy or finance may experience more scrutiny and external monitoring, leading to a stronger link between ESG performance and financial transparency (Cheng et al., 2014). Therefore, future research could also take into consideration the sector effects as a control variable in assessing the correlation between ESG and EQ.

Further research could be carried out by analysing firms from different countries to assess the influence of cultural and institutional factors, which may play a critical role. In civil-law countries with stronger regulatory frameworks and stakeholder-oriented governance (such as Italy), the deterrent effect of ESG compliance on earnings manipulation may be more pronounced (La Porta et al., 2000). On the contrary, in common-law jurisdictions with a stronger emphasis on shareholders, the effect might depend more heavily on market-based incentives.