Firm Performance and Corporate Social Responsibility: The Moderating Role of Board Skills

Abstract

Purpose: This study examines the association between firm profitability and corporate social responsibility (CSR), with a particular focus on the moderating role of board skills, specifically those with financial and industry expertise. Design/methodology/approach: Based on a sample of 42,623 observations from 2002 to 2021, we use panel regression analysis with robust standard errors are clustered at the firm level. Findings: The results show that firm profitability is positively associated with CSR performance. However, the positive effect is less likely in the presence of board with financial and industry expertise. Indeed, boards dominated by financially and industry experienced directors tend to prioritize short-term financial returns over long-term CSR initiatives. Originality/value: This study offers a novel contribution to stakeholder and legitimacy theory perspectives by show that the association between financial performance and CSR depends significantly on the expertise embedded within the boardroom. While financial and industry expertise can bring valuable oversight, it may not adequately support CSR initiatives. In this regard, firms may need directors with additional skills and perspectives—such as sustainability or stakeholder management expertise—to better address CSR issues and balance financial objectives with long-term societal and legitimacy concerns. Practical implications: Policy and decision makers should carefully consider the composition of the board when seeking to align profitability with corporate social responsibility (CSR) outcomes. While financial and industry expertise are valuable for oversight, an overrepresentation of such skills on the board may inadvertently undermine long-term CSR commitments.

1. Introduction

Corporate Social Responsibility (CSR) has transitioned from a peripheral activity to a central component of global business strategy (Morales-Parragué et al., 2022). Data from the OECD1 shows that as of 2025, the majority of the world’s largest companies now report on CSR metrics, with approximately 12,900 firms representing 91% of global market capitalization disclosing CSR-related data. This disclosure is largely driven by investor demand; according to Morgan Stanley’s 20252 “Sustainable Signals” report, 77% of global institutional investors say they actively integrate CSR factors into their investment decisions. Furthermore, 84% of investors state that CSR integration helps them identify higher-quality companies, and 64% believe it improves risk-adjusted returns. However, the global CSR landscape is becoming more complex. The OECD notes that while Europe leads with 98% of its major companies disclosing CSR data under regulations like the CSRD, the United States has seen political pushback, with some states passing anti-CSR legislation, creating a fragmented regulatory environment. Despite this political resistance in the US, global CSR assets under management remained above $30 trillion in 2025, and the majority of US institutional investors (60%) continue to consider CSR factors in their portfolios, suggesting that market forces are outpacing political opposition. Overall, while regulatory approaches to CSR vary significantly across jurisdictions, investor demand and market momentum remain the primary drivers of corporate CSR disclosure globally, indicating that CSR is no longer a choice but an expectation in modern capital markets.

The theoretical and empirical evidence regarding CSR is controversial. On the one hand, the agency theory suggests that CSR is embedded with high agency costs (Friedman, 1970; Jensen & Meckling, 2019). Specifically, managers and controlling shareholders may use firm resources to draw on the benefits of control (e.g., personal reputation) through CSR activities (Masulis & Reza, 2015; Barnea & Rubin, 2010). On the other hand, the stakeholder theory (Freeman, 1984) shows that CSR is value constructive for financial and non-financial stakeholders. CSR contributes to enhancing a firm’s value and competitiveness (Hakimi et al., 2025; Renneboog et al., 2008). It can benefit companies by offering strategic product market differentiation (Lins et al., 2017), higher financial performance (Hasan et al., 2018), and eventually, insurance against event risks (Godfrey et al., 2009).

In this context, a substantial body of studies has focused on whether CSR enhances firm performance, often framing CSR as a strategic investment capable of generating financial and competitive advantages (Gillan et al., 2021; Bahta et al., 2021). However, considerably less attention has been devoted to the reverse relationship, whether superior firm performance enables firms to engage more actively in CSR initiatives. This question is particularly important in an international context, where institutional environments, governance systems, and ESG reporting standards vary considerably across countries. Such variation provides an ideal setting for examining how financial success translates into CSR engagement under differing regulatory and stakeholder pressures. The association between firm performance and CSR is supported by the following theoretical perspectives. First, the resource-based view suggests that financially successful firms possess the slack resources necessary to support CSR investments (Uyar et al., 2024). While some CSR initiatives, such as environmental technology upgrades and sustainability infrastructure, require substantial financial commitments, others, including governance improvements, disclosure enhancements, and community engagement, are less capital-intensive (P. T. Lin et al., 2015). Second, stakeholder theory posits that highly profitable firms face greater scrutiny and heightened expectations from investors, customers, employees, and regulators (Aracil Fernández & Forcadell Martínez, 2024). Third, legitimacy theory suggests that firms use CSR to maintain social acceptance, protect corporate reputation, and reinforce their license to operate (Burlea & Popa, 2013). Together, these perspectives indicate that strong financial performance may both enable and motivate CSR engagement.

The primary objective of this paper is to examine the association between firm performance and CSR using an international panel of 42,623 firm-year observations from 2002 to 2021. We also investigate whether this relationship is moderated by board expertise, focusing specifically on directors’ financial and industry-specific skills (conceptual framework is presented in Figure 1). Our findings indicate that profitability positively influences CSR engagement, suggesting that more profitable firms have greater slack resources to allocate toward socially responsible initiatives. The positive association between firm performance and CSR is less likely in the presence of a board with financial and industry expertise. Boards dominated by directors with financial and industry backgrounds tend to focus more on firm short-term profitability, efficiency and operational resilience, which can weaken the positive influence of financial performance on CSR.

2. Literature Review and Hypothesis Development

Firm Profitability and Corporate Social Responsibility

While prior research has extensively examined whether corporate social responsibility (CSR) affects firm performance (Miller et al., 2020; Tang et al., 2012), considerably less attention has been devoted to the reverse relationship, whether superior financial performance enables firms to engage more actively in CSR initiatives. This reverse-causality perspective is particularly important because strong financial performance may either provide firms with the resources necessary to support CSR investments (Ruggiero & Cupertino, 2018) or, alternatively, reinforce managerial preferences for short-term financial objectives (Xu & Yang, 2023). Understanding these competing theoretical perspectives is therefore essential for explaining when and why profitable firms increase their commitment to CSR.

A substantial body of research has explored the CSR–performance link from a unidirectional perspective, emphasizing how CSR affects firm-level outcomes like profitability, innovation, and reputational capital (Hasan et al., 2018; Bahta et al., 2021). However, the reverse correlation, where firm performance acts as a driver of CSR engagement, has received comparatively less scholarly attention (Otero-González et al., 2021).

On one hand, drawing on the resource-based view, firms with strong financial performance are better positioned to engage in corporate social responsibility (CSR) because profitability generates slack resources, including surplus financial and human capital. These resources provide the flexibility needed to undertake social and environmental initiatives, particularly those requiring substantial long-term investment, without compromising core operations. Accordingly, slack resources constitute a key mechanism through which superior financial performance can translate into greater CSR engagement (Vitale et al., 2023).

In addition, stakeholder theory provides an additional explanation for why profitable firms may engage more actively in corporate social responsibility (CSR). As firms become more financially successful, they typically attract greater attention from investors, customers, employees, regulators, and other stakeholders. This heightened visibility often increases expectations for ethical conduct, social responsibility, and environmental stewardship (Forcadell et al., 2023). Moreover, stakeholder theory suggests that profitable firms are better positioned to address the increasing expectations of diverse stakeholders by undertaking substantive social and environmental initiatives (García-de-Madariaga & Rodríguez-de-Rivera-Cremades, 2010). Consequently, financially successful firms may expand their CSR activities to meet stakeholder demands and strengthen their relationships with key constituencies (Jamali, 2008).

In this context, CSR extends beyond a mere response to external pressures and becomes a strategic mechanism for preserving long-term legitimacy and organizational sustainability. CSR embodies both ethical responsibility and business opportunity, enabling firms to reconcile moral obligations with sustainable value creation (Kalra, 2024).

Complementing this perspective, the resource-based view suggests that profitable firms are better positioned to engage in CSR because superior financial performance generates slack resources that can be allocated to discretionary social and environmental initiatives (Hasan et al., 2018). These resources provide firms with the financial flexibility required to undertake CSR investments, particularly those involving substantial upfront costs and long-term returns.

Empirical evidence consistently supports this proposition. Firms with stronger financial performance, especially higher return on assets, tend to exhibit greater CSR engagement (Wang et al., 2016; Hasan et al., 2018; Swandari & Sadikin, 2016). Reduced financial constraints enable managers to invest more confidently in CSR initiatives, while stronger profitability also enhances firms’ willingness to disclose their social and environmental activities (Islam et al., 2021).

Taken together, both theoretical arguments and empirical findings indicate that firm profitability provides the resources, incentives, and strategic capacity necessary for meaningful CSR engagement. Accordingly, we propose the following hypothesis:

H1a. 

Firm profitability is positively associated with corporate social responsibility performance.

In contrast to the positive perspective, an alternative stream of research suggests that higher profitability may, under certain circumstances, reduce corporate social responsibility (CSR) engagement. Financial success does not automatically translate into greater social investment, as CSR decisions are shaped not only by resource availability but also by managerial incentives, shareholder expectations, and strategic priorities (Babalola, 2012).

Agency theory offers a key explanation for this negative relationship. Shareholders may view CSR as a discretionary expenditure that diverts resources from value-maximizing activities (Borghesi et al., 2014). When firms generate strong profits, investors often prefer that excess cash be distributed through dividends, share repurchases, or reinvestment in core operations rather than allocated to CSR initiatives (Y. Zhang & Gimeno, 2016). This pressure can encourage managers to limit CSR spending, particularly when its financial benefits are uncertain.

Short-term financial priorities further reinforce this tendency. Managers of highly profitable firms frequently face pressure to sustain earnings growth and deliver immediate returns. Because many CSR initiatives require substantial upfront investment while generating benefits only over the long term, they may be viewed as less attractive than projects with quicker and more measurable financial payoffs (Charlo et al., 2017). Consequently, firms may prioritize short-term profitability over broader social and environmental commitments.

This argument is also consistent with Jensen’s free cash flow theory (Jensen, 1986), which suggests that excess financial resources can intensify agency conflicts. In such situations, CSR may be perceived as a non-essential expenditure or as a managerial tool that does not necessarily maximize shareholder value (Barnea & Rubin, 2010). Moreover, highly profitable firms may face lower legitimacy pressures, reducing the strategic need to invest heavily in CSR (Aragón-Correa & Rubio-López, 2007).

Taken together, these arguments indicate that profitability may, in some cases, discourage CSR engagement rather than promote it. Accordingly, we propose the following competing hypothesis:

H1b. 

Firm profitability is negatively associated with corporate social responsibility performance.

3. Research Design

3.1. Sample Selection and Data

This study employs an international panel dataset comprising 42,623 firm-year observations from 57 countries over the period 2002–2021. The sample period begins in 2002, when the necessary ESG data first became consistently available, and ends in 2021, the latest reporting year available at the time of data collection. The initial sample includes all publicly listed non-financial firms with available data in both Compustat and Refinitiv Eikon. Firm-level financial and accounting data were obtained from Compustat, while corporate social responsibility (CSR) was measured using the Asset4 Environmental, Social, and Governance (ESG) data available from the Refinitiv Eikon database. Following prior literature, firms in the financial sector were excluded because their distinct regulatory environment and accounting practices make them less comparable to non-financial firms. The two databases were matched using unique firm identifiers. Duplicate observations were removed, and firm-year observations with missing values for the dependent, independent, or control variables were excluded. All continuous variables were winsorized at the 1st and 99th percentiles to mitigate the influence of extreme values and potential outliers. This standard procedure reduces the impact of unusually large observations while preserving the overall distribution of the data. The final sample consists of 42,623 firm-year observations across nine Industry Classification Benchmark (ICB) sectors.

Panel A of

Table 1. Sample distribution.

Panel A. Sample Distribution per Country
Country	N	Percentage
Argentina	86	0.202
Australia	1937	4.544
Austria	187	0.439
Belgium	278	0.652
Bermuda	143	0.335
Brazil	106	0.249
Canada	1954	4.584
Cayman Islands	42	0.099
Chile	226	0.530
China	3235	7.590
Colombia	45	0.106
Cyprus	32	0.075
Denmark	328	0.770
Egypt	48	0.113
Finland	388	0.910
France	1004	2.356
Germany	1149	2.696
Greece	126	0.296
Hong Kong	1145	2.686
Hungary	38	0.089
India	903	2.119
Indonesia	326	0.765
Ireland	383	0.899
Isle of Man	11	0.026
Italy	360	0.845
Japan	3429	8.045
South Korea	734	1.722
Kuwait	28	0.066
Luxembourg	139	0.326
Macau	33	0.077
Malaysia	510	1.197
Mexico	300	0.704
Morocco	10	0.023
Netherlands	465	1.091
New Zealand	318	0.746
Norway	280	0.657
Oman	13	0.030
Panama	13	0.030
Peru	99	0.232
Philippines	200	0.469
Poland	215	0.504
Portugal	81	0.190
Qatar	11	0.026
Russia	293	0.687
Saudi Arabia	99	0.232
Singapore	545	1.279
South Africa	859	2.015
Spain	301	0.706
Sweden	925	2.170
Switzerland	874	2.051
Thailand	405	0.950
Turkey	102	0.239
Ukraine	11	0.026
United Arab Emirates	22	0.052
United Kingdom	3403	7.984
United States of America	13,426	31.499
Total	42,623	100.000
Panel B. Sample Distribution per Year
Year	N	Percentage
2002	89	0.209
2003	179	0.420
2004	346	0.812
2005	490	1.150
2006	537	1.260
2007	836	1.961
2008	1145	2.686
2009	1353	3.174
2010	1608	3.773
2011	1732	4.064
2012	2071	4.859
2013	2228	5.227
2014	2265	5.314
2015	2755	6.464
2016	3215	7.543
2017	3818	8.958
2018	4353	10.213
2019	5130	12.036
2020	5743	13.474
2021	2730	6.405
Total	42,623	100.000
Panel C. Sample Distribution per Industry
Industry	N	Percentage
Basic Materials	5208	12.219
Consumer Cyclicals	8056	18.901
Consumer Non-Cyclicals	4104	9.629
Energy	3063	7.186
Healthcare	3668	8.606
Industrials	8436	19.792
Real Estate	2433	5.708
Technology	6818	15.996
Utilities	837	1.964
Total	42,623	100.000

displays the geographical composition of the sample dataset. U.S.-based entities constitute the largest proportion at 31.49%, followed by Japanese firms at 8.04%. Ukrainian companies represent the smallest cohort with a 0.026% share. Panel B presents the temporal distribution, revealing an upward trend from 0.209% in 2002 to a peak of 13.474% in 2020. Regarding the sector-level distributions, Panel C of Table 1 reports the ratios of the sample ranged between 1.964% (Utilities) and 19.792% (Industrials).

3.2. Variable Definitions

3.2.1. Corporate Social Responsibility

Our corporate social responsibility (CSR) is measured using Refinitiv Eikon’s overall ESG score, which captures firms’ environmental, social, and governance performance (Govindan et al., 2021). The environmental dimension incorporates performance metrics related to eco-innovation, emission reduction, and resource efficiency. The social component assesses workforce conditions, human rights compliance, community engagement, and product safety standards. The governance dimension evaluates board management quality, shareholder rights protection, and strategic CSR integration. Because the governance dimension may overlap conceptually with governance-related explanatory variables, we conduct an additional robustness test using the combined environmental and social (ES) score, excluding the governance component. The results remain qualitatively unchanged, suggesting that our findings are not driven by mechanical overlap between the dependent and independent variables.

3.2.2. Firm Performance

Our independent variable is firm profitability, measured by Return on Assets (ROA). Consistent with (Utomo & Simanungkalit, 2024), ROA is calculated using the following formula:

$$\mathrm{ROA}={\displaystyle \frac{\mathrm{I}\mathrm{n}\mathrm{c}\mathrm{o}\mathrm{m}\mathrm{e} \mathrm{b}\mathrm{e}\mathrm{f}\mathrm{o}\mathrm{r}\mathrm{e} \mathrm{t}\mathrm{a}\mathrm{x}}{\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{A}\mathrm{s}\mathrm{s}\mathrm{e}\mathrm{t}\mathrm{s}}}$$

According to Astuti et al. (2025), Return on Assets is widely recognized as a key indicator of how efficiently management converts company assets into earnings, making it a crucial measure for assessing corporate performance and success.

3.2.3. Control Variables

Consistent with prior research (e.g., Wang et al., 2016; Barnea & Rubin, 2010; Chen et al., 2021; Smith & Huang, 2023), we include several firm- and country-level control variables that may influence CSR engagement. LEVERAGE is measured as total debt divided by total assets. A negative relationship is expected, as highly leveraged firms face greater financial constraints and may have fewer slack resources available for discretionary CSR activities (Simionescu & Gherghina, 2014). FIRM_SIZE is calculated as the natural logarithm of total assets. Larger firms are expected to exhibit higher CSR engagement because they possess greater resources and face increased stakeholder scrutiny (Hossain & Karim, 2023).

CEO_DUALITY is a dummy variable equal to one if the chief executive officer also serves as board chair, and zero otherwise. A negative association is anticipated, as CEO duality may weaken board independence and reduce oversight of CSR-related decisions (Karagiannopoulou et al., 2026). BOARD_GENDER is measured as the proportion of female directors on the board. A positive effect is expected because gender-diverse boards often enhance monitoring quality and promote stronger attention to social and environmental issues (Zahid et al., 2025). BOARD_SIZE represents the total number of directors on the board. Larger boards may provide broader expertise and stronger oversight, although excessively large boards may also suffer from coordination challenges (Zahid et al., 2025).

GDP is measured as the natural logarithm of gross domestic product per capita. Firms operating in more developed economies are generally expected to demonstrate stronger CSR engagement due to greater institutional development and stakeholder awareness. Finally, AUDIT_COMMITTEE and GOVERNANCE_COMMITTEE are binary variables indicating the existence of an audit committee and a corporate governance committee, respectively. Positive coefficients are expected for both variables, as these governance mechanisms enhance oversight, transparency, and alignment with stakeholder interests, thereby supporting stronger CSR engagement (Cavaco & Crifo, 2010; L. Zhang & Su, 2023).

3.3. Model Specification and Empirical Analysis

To examine the effect of firm profitability on corporate social responsibility, following the ordinary least squares (OLS) method using this model:

CSR_it = α₀ + α₁ ROA_it + α₂ LEVERAGE_it + α₃ FIRM_SIZE_it + α₄ CEO_DUALITY_it + α₅ BOARD_GENDER_it + α₆ BOARD_SIZE_it + α₇ GDP_it + α₈ AUDIT_COMMITTEE_it + α₉ GOVERNANCE_COMMITTEE_it + COUNTRY FIXED EFFECT + YEAR FIXED EFFECT + INDUSTRY FIXED EFFECT + ε_it

where corporate social responsibility is the dependent variable proxied by the ESG score and denotes CSR. Firm profitability, measured by return on assets (ROA), serves as the primary independent variable. The analysis also includes several control variables: firm leverage (LEVERAGE), firm size (FIRM_SIZE), CEO duality (CEO_DUALITY), board gender diversity (BOARD_GENDER), board size (BOARD_SIZE), gross domestic product (GDP), the presence of an audit committee (AUDIT_COMMITTEE), and the presence of a governance committee (GOVERNANCE_COMMITTEE).

To examine the relationship between firm profitability and corporate social responsibility, this study employs panel data regression models with firm-level observations across countries and years. Consistent with prior literature, all estimations include country, year, and industry fixed effects to control for time-invariant country characteristics, global temporal shocks, and sector-specific heterogeneity. By accounting for these sources of unobserved heterogeneity, the fixed-effects specification helps reduce omitted variable bias and strengthen the internal validity of the estimated relationships. Robust standard errors are clustered at the firm level to account for serial correlation and heteroscedasticity within firms over time, thereby ensuring reliable statistical inference.

3.4. Descriptive Statistics

Table 2. Descriptive Statistics.

Variable	Obs	Mean	Std. Dev.	Min	Max
CSR	42,623	43.918	19.837	12.806	78.828
ROA	42,623	0.067	0.078	−0.099	0.230
LEVERAGE	42,623	0.244	0.166	0.000	0.574
FIRM_SIZE	42,623	21.501	1.561	18.471	24.245
CEO_DUALITY	42,623	0.368	0.482	0.000	1.000
BOARD_GENDER	42,623	15.862	12.525	0.000	40.000
BOARD_SIZE	42,623	9.711	3.167	1.000	33.000
GDP	42,623	10.481	0.727	8.733	11.160
AUDIT_COMMITTEE	42,623	0.917	0.276	0.000	1.000
GOVERNANCE_COMMITTEE	42,623	0.459	0.498	0.000	1.000

This table presents the summary statistics for all variables used in the regression analysis based on a sample of 42,623 firm-year observations over the period 2002–2021. Continuous variables are winsorized at the 1st and 99th percentiles to mitigate the influence of outliers. Detailed variable definitions are provided in Appendix A.

presents summary statistics for all variables included in the primary analysis, based on a sample of 42,623 observations spanning the period 2002–2021. Robust standard errors are clustered at the firm level to account for serial correlation and heteroscedasticity within firms over time, thereby ensuring reliable statistical inference. Prior to estimation, all continuous variables are winsorized at the 1st and 99th percentiles to mitigate the influence of extreme observations and outliers. In terms of firm profitability, the average value of return on assets (ROA) was 0.067. Also, about the dependent variable, the corporate social responsibility (CSR) average value is 43.918. The descriptive statistics results for all other controls generally align with findings from previous studies (Otero-González et al., 2021).

3.5. Correlation Matrix

Table 3. Correlation Matrix.

Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
(1) CSR	1.000
(2) ROA	0.029 ***	1.000
(3) LEVERAGE	0.086 ***	−0.276 ***	1.000
(4) FIRM_SIZE	0.441 ***	0.105 ***	0.134 ***	1.000
(5) CEO_DUALITY	−0.054 ***	0.043 ***	−0.007	0.116 ***	1.000
(6) BOARD_ GENDER	0.312 ***	−0.015 ***	0.062 ***	0.016 ***	−0.015 ***	1.000
(7) BOARD_SIZE	0.271 ***	0.024 ***	0.108 ***	0.460 ***	0.057 ***	0.007	1.000
(8) GDP	0.042 ***	−0.088 ***	−0.008 *	0.018 ***	0.141 ***	0.175 ***	−0.137 ***	1.000
(9) AUDIT_COMMITTEE	0.081 ***	−0.007	0.067 ***	−0.051 ***	0.091 ***	0.179 ***	−0.015 ***	0.097 ***	1.000
(10) GOVERNANCE_COMMITTEE	0.023 ***	−0.048 ***	0.062 ***	0.079 ***	0.290 ***	0.107 ***	0.025 ***	0.309 ***	0.253 ***	1.000
Mean VIF	1.18
VIF		1.12	1.13	1.34	1.12	1.07	1.31	1.18	1.11	1.26

This table presents the correlation matrix for all regression variables based on 42,623 firm-year observations from 2002 to 2021. Statistical significance at the 1% and 10% levels is denoted by *** and * respectively. Complete variable definitions are provided in Appendix A.

displays the Pearson correlation matrix for the dependent, independent, and control variables. The correlation matrix and variance inflation factor (VIF) analyses are computed using the final winsorized estimation sample. All correlation coefficients remain below the 0.8 threshold. This finding is further supported by variance inflation factor (VIF) values ranging from 1.07 to 1.34, well below the critical value of 10. The results indicate no evidence of severe multicollinearity among the explanatory variables, supporting the reliability of the regression estimates. This empirical framework provides a rigorous basis for assessing the relationship between profitability and CSR engagement.

4. Empirical Analysis

This section set out to expose the sample empirical analysis, additional tests and robustness check of the findings to test the connection between profitability and corporate social responsibility, as well as the board skills moderator effect.

4.1. The Effect of Firm Performance on Corporate Social Responsibility

The findings presented in

Table 4. Firm Profitability and Corporate Social Responsibility

	CSR
ROA	4.485 ***
	(4.409)
LEVERAGE	−1.448 ***
	(−2.951)
FIRM_SIZE	6.666 ***
	(108.865)
CEO_DUALITY	−1.525 ***
	(−9.073)
BOARD_GENDER	0.322 ***
	(43.409)
BOARD_SIZE	0.384 ***
	(13.161)
GDP	3.339 ***
	(4.111)
AUDIT_COMMITTEE	6.637 ***
	(20.060)
GOVERNANCE_COMMITTEE	5.967 ***
	(25.001)
CONSTANT	−151.841 ***
	(−19.864)
COUNTRY	Yes
INDUSTRY	Yes
YEAR	Yes
OBSERVATIONS	42,623
R_SQUARED	0.445

This table reports the effect of firm profitability on corporate social responsibility (CSR). Profitability is measured by return on assets (ROA). The sample consists of 42,623 firm-year observations from 2002 to 2021. t-statistics are reported in parentheses. *** indicates statistical significance at the 1% levels. Variable definitions are provided in Appendix A.

provide robust empirical support for a positive association between firms’ financial performance and corporate social responsibility practices. Specifically, the results shown in Table 4 demonstrate that higher return on assets (ROA), a widely recognized indicator of firm profitability, is significantly and positively associated with enhanced CSR performance. Economically, the estimated coefficient suggests that a one-standard-deviation increase in profitability is associated with an approximately 0.35 increase in CSR engagement, after controlling for the full set of firm-level, governance, and country-level variables. This result indicates that the positive association is not only statistically significant but also economically meaningful, thereby confirming our first hypothesis (H1a). This outcome suggests that firms with stronger financial performance possess greater capacity to allocate resources toward social and environmental initiatives.

From a theoretical perspective, this positive association is consistent with the resource-based view, which posits that profitable firms generate slack resources that can be allocated to discretionary activities such as corporate social responsibility. Greater financial flexibility enables firms to undertake social and environmental initiatives without compromising their core operations.

This finding is also aligned with stakeholder theory. Financially successful firms typically attract greater attention from investors, customers, employees, regulators, and local communities, thereby increasing expectations for responsible corporate behavior. In response, profitable firms may engage more actively in CSR to strengthen stakeholder relationships, enhance corporate legitimacy, and preserve their long-term license to operate.

Overall, our findings indicate that more profitable firms tend to exhibit higher levels of CSR engagement. This result supports the view that financial performance provides both the resources and the strategic incentives necessary for meaningful CSR investment. It is also consistent with prior studies (including Otero-González et al., 2021), which document that financially stronger firms are generally more inclined to undertake socially responsible activities.

4.2. Additional Analysis: The Moderating Role of Board Skills

Board skills play a critical role in shaping corporate social responsibility (CSR) performance (C. C. Lin & Nguyen, 2022). Directors with industry-specific expertise are better equipped to understand sector-specific risks, regulatory challenges, and stakeholder expectations related to social and environmental issues, thereby potentially enhancing the effectiveness of CSR initiatives (Shafeeq Nimr Al-Maliki et al., 2023). Similarly, financial expertise strengthens board oversight by enabling directors to evaluate the economic implications of CSR investments and assess whether such initiatives are aligned with long-term value creation for both shareholders and other stakeholders (Pahuja & Agrawal, 2023). Accordingly, boards possessing substantial financial and industry expertise are better positioned to process complex information, make informed strategic decisions, and engage effectively with diverse stakeholders, all of which may enhance CSR performance (C. C. Lin & Nguyen, 2022).

Conversely, agency theory suggests that while boards possessing financial and industry expertise are often presumed to enhance strategic decision-making, such competencies may paradoxically bias managerial attention toward shareholder value maximization (Barnea & Rubin, 2010). This perspective is further supported by the managerial opportunism hypothesis (Prior et al., 2008), which posits that managers may sometimes advance firm short-term profitability over long-term CSR initiatives (Barnea & Rubin, 2010; Prior et al., 2008).

In this study, we argue that the positive effect of CSR on firm performance is moderated by board skills. Following Nizar et al. (2025), board skills are measured as the percentage of directors possessing either industry-specific expertise or financial expertise. Industry expertise refers to directors with substantial professional or board experience in the firm’s primary industry, while financial expertise captures directors with backgrounds in accounting, finance, banking, or related fields.

The results reported in

Table 5. The role of Board Skills.

	CSR
ROA	9.361 ***
	(3.625)
BOARDSKILLS	0.033 ***
	(6.582)
ROA X BOARDSKILLS	−0.093 **
	(−2.098)
LEVERAGE	−1.368 ***
	(−2.788)
FIRM_SIZE	6.680 ***
	(109.094)
CEO_DUALITY	−1.574 ***
	(−9.362)
BOARD_GENDER	0.324 ***
	(43.692)
BOARD_SIZE	0.398 ***
	(13.609)
GDP	3.130 ***
	(3.852)
AUDIT_COMMITTEE	6.637 ***
	(20.071)
GOVERNANCE_COMMITTEE	5.954 ***
	(24.959)
CONSTANT	−151.299 ***
	(−19.801)
COUNTRY	Yes
INDUSTRY	Yes
YEAR	Yes
OBSERVATIONS	42,623
R-SQUARED	0.446

This table reports the moderating effect of board competencies on the relationship between firm profitability and corporate social responsibility (CSR). The sample consists of 42,623 firm-year observations from 2002 to 2021. t-statistics are reported in parentheses. *** and ** indicate statistical significance at the 1% and 5% levels, respectively. Variable definitions are provided in Appendix A.

indicate that board skills moderate the relationship between firm profitability and CSR engagement. Although both profitability and board skills exhibit positive direct associations with CSR, the negative and statistically significant interaction term suggests that board expertise attenuates the positive profitability-CSR association. This finding suggests that while financial and industry expertise are important, they may not be sufficient to fully support CSR. Boards dominated by directors with financial and industry backgrounds tend to focus more on firm short-term profitability, efficiency and operational resilience, which can weaken the positive influence of financial performance on CSR. This implies that firms facing complex social, environmental, and stakeholder-related challenges require a broader mix of board competencies. Firms may need directors with additional skills and perspectives—such as sustainability or stakeholder management expertise—to better address CSR issues and balance financial objectives with long-term societal and legitimacy concerns. This interpretation should be viewed as theoretically grounded rather than conclusive. The logic to this is that the strength of this moderating effect may vary across institutional environments, where governance practices and stakeholder expectations differ.

4.3. Robustness Check

4.3.1. Alternative Metrics of Firm Performance

To ensure the robustness of our results, we re-estimate our core model using three different profitability metrics Semykina and Wooldridge (2010). A robust finding should hold across conceptually related but distinct measures of a company’s profitability level. This test helps to alleviate concerns that the findings are attributable to measurement error or peculiarities inherent to a specific accounting metric. In

Table 6. Alternative Metrics of Firm Performance.

	(1)	(2)	(3)
	CSR	CSR	CSR
ROE	3.808 ***
	(4.229)
NET_PROFIT_MARGIN		14.754 ***
		(17.942)
TOBINQ			0.151 ***
			(4.730)
LEVERAGE	−1.798 ***	−0.579	−1.790 ***
	(−3.785)	(−1.215)	(−3.773)
FIRM_SIZE	6.668 ***	6.746 ***	6.737 ***
	(108.938)	(111.316)	(109.935)
CEO_DUALITY	−1.517 ***	−1.553 ***	−1.518 ***
	(−9.029)	(−9.278)	(−9.036)
BOARD_GENDER	0.322 ***	0.317 ***	0.322 ***
	(43.416)	(42.860)	(43.377)
BOARD_SIZE	0.383 ***	0.360 ***	0.380 ***
	(13.126)	(12.378)	(13.008)
GDP	3.286 ***	3.524 ***	3.191 ***
	(4.047)	(4.356)	(3.930)
AUDIT_COMMITTEE	6.633 ***	6.803 ***	6.598 ***
	(20.048)	(20.637)	(19.957)
GOVERNANCE_COMMITTEE	5.962 ***	6.033 ***	5.924 ***
	(24.983)	(25.378)	(24.836)
CONSTANT	−151.198 ***	−155.938 ***	−151.549 ***
	(−19.783)	(−20.464)	(−19.829)
COUNTRY	Yes	Yes	Yes
INDUSTRY	Yes	Yes	Yes
YEAR	Yes	Yes	Yes
OBSERVATIONS	42,623	42,623	42,623
R_SQUARED	0.445	0.449	0.446

This table presents the primary findings examining the relationship between firm profitability and corporate social responsibility (CSR) using multiple profitability metrics: return on equity (ROE), net profit margin, and Tobin’s Q. The analysis employs a sample of 42,623 firm-year observations spanning the period 2002–2021. Statistical significance at the 1% levels is denoted by ***. Comprehensive variable definitions are provided in Appendix A.

, we replace our primary measure, return on assets (ROA), with first the return on equity (ROE), which is calculated as the ratio of net income to shareholders’ equity (Jihadi et al., 2021). ROE shifts the focus from the efficiency of all assets to the return generated specifically for equity shareholders, capturing a different dimension of profitability. Second the Net profit margin (NPM) is calculated as the ratio of net income to net sales revenue (Khabibah et al., 2025). NPM assesses a firm’s operational efficiency by showing the percentage of revenue that translates into profit, independent of the firm’s asset base or leverage. Third, Tobin’s Q is employed as a market-based performance measure that captures firm valuation, investor expectations, and future growth opportunities, thereby complementing the accounting-based profitability measures used in the baseline analysis (Farooq & Ahmad, 2023; Astuti et al., 2025).

The robust results with Tobin’s Q are particularly noteworthy. They suggest that the positive relationship holds not only for traditional accounting measures (ROA, ROE, and NPM) but also for a forward-looking market-based measure. This implies that the market’s anticipation of future earnings is also associated with higher CSR commitment. The fact that all three alternative proxies confirm our core finding significantly reinforces the robustness of the positive effect that firm profitability has on CSR commitments.

4.3.2. Alternative Metrics of CSR

To further assess the robustness of our findings, we examine whether the positive relationship between firm profitability and corporate social responsibility extends to alternative CSR dimensions. Specifically, we re-estimate our baseline model using four alternative dependent variables: Environmental Pillar Score, Social Pillar Score, Governance Pillar Score, and the combined Environmental and Social (ES) Performance score. As reported in

Table 7. Alternative CSR metrics.

	(1)	(2)	(3)	(4)
	Envir_Pillar_Score	Social_Pillar_Score	Governance_Pillar_Score	ES_Perf
ROA	8.505 ***	4.561 ***	8.398 ***	2.782 **
	(7.496)	(3.753)	(6.615)	(2.365)
LEVERAGE	1.443 **	−1.576 ***	−1.320 **	−1.984 ***
	(2.565)	(−2.638)	(−2.108)	(−3.386)
FIRM_SIZE	−5.060 ***	7.398 ***	4.328 ***	8.304 ***
	(−70.055)	(96.895)	(55.638)	(109.268)
CEO_DUALITY	4.177 ***	0.419 **	−6.122 ***	0.583 ***
	(21.214)	(2.016)	(−28.310)	(2.832)
BOARD_GENDER	−0.400 ***	0.255 ***	0.467 ***	0.269 ***
	(−46.442)	(27.946)	(49.432)	(30.460)
BOARD_SIZE	0.363 ***	0.641 ***	−0.618 ***	0.800 ***
	(10.030)	(16.451)	(−16.229)	(20.668)
GDP	−1.363	−2.652 ***	7.408 ***	1.875 *
	(−1.408)	(−2.630)	(7.156)	(1.889)
AUDIT_COMMITTEE	−10.736 ***	5.801 ***	11.571 ***	4.696 ***
	(−26.488)	(13.204)	(27.423)	(10.798)
GOVERNANCE_COMMITTEE	−10.758 ***	4.974 ***	11.751 ***	3.967 ***
	(−38.804)	(16.987)	(39.083)	(14.212)
CONSTANT	−77.015 ***	−130.052 ***	−99.313 ***	191.417 ***
	(−8.374)	(−13.694)	(−10.116)	(−20.699)
COUNTRY	Yes	Yes	Yes	Yes
INDUSTRY	Yes	Yes	Yes	Yes
YEAR	Yes	Yes	Yes	Yes
OBSERVATIONS	42,623	42,623	42,623	42,623
R-SQUARED	0.298	0.419	0.238	0.466

This table presents additional analyses examining the relationship between firm profitability and alternative dimensions of corporate social responsibility (CSR), including Environmental Pillar Score, Social Pillar Score, Governance Pillar Score, and the combined Environmental and Social (ES) Performance score. Firm profitability is measured by return on assets (ROA). The analysis is based on a sample of 42,623 firm-year observations covering the period 2002–2021. Statistical significance at the 1%, 5%, and 10% levels is denoted by ***, **, and *, respectively. All model specifications include the full set of control variables, as well as year, industry, and country fixed effects. Standard errors are clustered at the firm level. Detailed variable definitions are provided in Appendix A.

, firm profitability remains positively and statistically significantly associated with all four CSR measures. These results indicate that the positive association of profitability is not confined to the aggregate ESG score, but is consistently observed across its individual pillars and the non-governance dimensions. Collectively, these findings reinforce our main conclusions and provide further evidence that more profitable firms are better positioned to invest in a broad range of sustainability and social responsibility initiatives.

4.3.3. Endogeneity Concerns

Entropybalancing

To mitigate endogeneity concerns arising from selection bias and observable differences between treated and control firms, we employ the entropy balancing technique (Hainmueller, 2012). This method re-weights the control group to achieve perfect balance on a set of pre-treatment covariates, ensuring that the groups are comparable on these observed dimensions. This approach improves covariate balance and mitigates bias arising from observable firm characteristics. However, as with other matching-based techniques, it cannot fully eliminate concerns related to unobserved heterogeneity or reverse causality. Compared to propensity score matching, entropy balancing achieves superior covariate balance across key statistical moments, including the mean, variance, and skewness. Following the protocol established by Hainmueller (2012), we confirm convergence for the balanced covariates across all three moments. The multivariate regression results reported in

Table 8. Robustness to Endogeneity

	Entropy Balancing	2SLS
	CSR	ROA	CSR
ROA	4.080 ***		8.822 ***
	(3.847)		(2.952)
L.ROA		0.685 ***
		(97.728)
PEER_ROA		0.536 ***
		(20.688)
LEVERAGE	1.801 ***	−0.053 ***	−0.387
	(2.899)	(−21.746)	(−0.315)
FIRM_SIZE	6.298 ***	0.003 ***	6.576 ***
	(84.990)	(12.166)	(39.240)
CEO_DUALITY	−1.292 ***	−0.000	−1.617 ***
	(−6.490)	(−0.040)	(−3.933)
BOARD_GENDER	0.305 ***	0.000	0.330 ***
	(34.611)	(1.408)	(19.622)
BOARD_SIZE	0.478 ***	−0.000 ***	0.346 ***
	(11.869)	(−4.003)	(4.257)
GDP	2.957 ***	−0.008 **	4.168 ***
	(3.072)	(−2.432)	(2.774)
AUDIT_COMMITTEE	7.085 ***	−0.003 ***	6.855 ***
	(16.199)	(−2.780)	(7.659)
GOVERNANCE_COMMITTEE	6.069 ***	−0.003 ***	5.941 ***
	(21.461)	(−2.905)	(10.088)
CONSTANT	−141.612 ***	0.003	−144.239 ***
	(−15.779)	(0.085)	(−9.913)
COUNTRY	Yes	Yes	Yes
INDUSTRY	Yes	Yes	Yes
YEAR	Yes	Yes	Yes
OBSERVATIONS	42,623	35,316	35,316
R-SQUARED	0.446	0.594	0.438
UNDERID. (LM)			1449.78 ***
WEAK ID. (CRAGG-DONALD WALD F)			5812.31

Table 8 presents robustness checks addressing potential endogeneity concerns in the profitability-CSR relationship, including omitted variable bias and reverse causality. The analysis employs a sample of 42,623 firm-year observations spanning the period 2002–2021. Statistical significance at the 1% and 5% levels is denoted by *** and **, respectively. Comprehensive variable definitions are provided in Appendix A.

reinforce our primary findings, demonstrating a consistent and statistically significant positive impact of a company’s profitability on corporate social responsibility (CSR). This robustness check substantiates our main conclusions by showing that the association holds regardless of the balancing method employed.

Two-Stage Least Squares (2SLS)

To further address endogeneity concerns, we complement the entropy balancing analysis with an instrumental variable approach based on two-stage least squares (2SLS). While entropy balancing mitigates selection bias arising from observable firm characteristics, it cannot fully eliminate endogeneity stemming from unobserved heterogeneity, reverse causality, or simultaneity. Therefore, employing both approaches provides a more comprehensive assessment of the robustness of our findings. Specifically, we instrument firm profitability (ROA) using two widely accepted instruments: the one-period lag of ROA and the average profitability of peer firms within the same industry-year, excluding the focal firm. The lagged ROA captures the persistence of firm performance while remaining predetermined with respect to current CSR decisions. Peer profitability reflects common industry-level economic conditions that affect a firm’s profitability but are unlikely to directly influence its CSR engagement beyond their effect on firm profitability. The diagnostic tests strongly support the validity and relevance of the instruments. First, the underidentification test is highly significant (LM = 1449.78, p < 0.001), confirming that the instruments are strongly correlated with firm profitability. Second, the weak identification statistic is 5812.31, which far exceeds the Stock-Yogo critical value of 19.93, indicating that weak instruments are not a concern. Finally, the Durbin-Wu-Hausman test rejects the null hypothesis that profitability is exogenous (F = 14.58, p < 0.001), confirming the appropriateness of the instrumental variable approach. Together, these results provide strong support for the validity and strength of the selected instruments. The 2SLS results, reported in Table 8 (Columns 2 and 3), remain fully consistent with our baseline and entropy balancing estimates. Firm profitability continues to exhibit a positive and statistically significant effect on CSR.

Taken together, both the entropy balancing and instrumental variable analyses provide compelling evidence that our main findings are robust to both observable and unobservable sources of endogeneity, thereby strengthening the causal interpretation of the positive relationship between profitability and CSR.

5. Conclusions

The primary aim of this research is to examine the impact of firm financial performance on corporate social responsibility (CSR), providing insights into how profitable firms develop and sustain CSR practices over time and across countries. Using an international sample of 42,623 firm-year observations from 2002 to 2021, we find that stronger financial performance is associated with higher levels of CSR engagement. This result suggests that profitable firms possess greater financial flexibility and organizational resources, enabling them to invest more extensively in socially responsible activities. The finding is consistent with stakeholder theory, which posits that highly profitable firms attract greater scrutiny and expectations from investors, customers, employees, regulators, and society at large. Consequently, financially successful firms are better positioned, and often more motivated, to meet these expectations through sustained CSR engagement.

Additional analyses reveal that the positive relationship between profitability and CSR becomes weaker in firms with boards possessing greater financial and industry expertise. Specifically, the negative interaction term indicates that board skills influence how profitability is translated into CSR investment.

Our findings carry several important implications. First, firms seeking to strengthen their CSR engagement may benefit from appointing directors with broader perspectives, especially those with expertise in sustainability, ethics, or stakeholder management. Nomination board committee may benefit from incorporating members with broader expertise (e.g., sustainability expertise), as this can enhance the quality of strategic oversight, strengthen consideration of sustainability and ethical issues, and improve alignment between corporate strategy and CSR objectives. Policy makers interested in promoting CSR may consider encouraging or requiring more diverse board expertise, or designing governance guidelines that support broader stakeholder-focused decision-making. Managers should provide board members with financial and industry expertise, evidence-based business cases showing how CSR initiatives drive long-term value and strengthen risk management. This helps shift their attention away from narrow short-term performance objectives toward a more balanced and responsible strategic perspective.

This research is not free from limitations. First, the sample is unbalanced across countries, industries, and years, reflecting differences in data availability and disclosure practices. Second, although our empirical approach incorporates extensive controls and multiple robustness tests, including entropy balancing and additional endogeneity analyses, the study does not permit definitive causal inference, and reverse causality cannot be entirely ruled out. Future research could extend our study by examining the moderating role of other dimensions of board expertise, such as sustainability, technological, or international experience. These forms of expertise may further influence how boards’ skills navigate the tension between shareholder value maximization and broader stakeholder accountability.