This study aimed to investigate the role of the country's institutional quality on the environmental, social and governance (ESG) performance of its companies.
Over a four-year period (2016–2019), the study examined the ESG performance of 412 organizations situated in 19 countries. ESG performance was the dependent variable, and the independent variables were rule of law, economic freedom, education index and international trade freedom. These factors described the institutional quality of countries in the authors’ study.
The findings reveal that institutional quality has a major impact on ESG performance. Companies engage in more ESG practices when they operate in countries with greater economic freedom and international trade freedom. The authors corroborated the core assumption of institutional theory (IT), which argues that organizational behavior is determined by the country's institutional setting.
The findings, like all research, should be interpreted with caution. The authors’ research focused solely on large energy corporations. As a result, the conclusions cannot be applied to small companies or other industries. ESG performance can also be measured using different datasets.
If managers want their companies to perform better in terms of ESG, the authors recommend that they form a CSR committee and sign the Global Compact. This study may be valuable to international policymakers because they can underline that greater economic freedom, better education and greater international trade freedom all promote higher ESG performance.
To the best of the authors' knowledge, nearly all of research explores the relationship between ESG and financial performance. As a result, this study built on past research by investigating how national aspects affect corporate ESG performance.
1. Introduction
Environmental, social and governance (ESG) performance has begun to emerge in studies of corporate sustainability (Ortas et al., 2019). ESG performance focuses on stakeholder-oriented management, which means that social and environmental information disclosed by companies is of interest not only to shareholders, but also to consumers, clients, communities, media, the State, among other interested parties (Abdi et al., 2022). From this perspective, ESG performance differs from traditional shareholder-oriented management as it identifies three forms of stakeholder relationships: environmental, social and governance (Ortas et al., 2019).
ESG data is a non-financial disclosure that does not follow a conventional structure like financial data (Buallay, 2019). The ESG performance of corporations in the energy industry is critical since this segment directly consumes natural resources while also emitting the most CO2 into the environment (Pinheiro et al., 2023). As a result, society expects better levels of ESG performance (Lu et al., 2019). Furthermore, ESG performance has been critical to the survival of energy companies, as investors are interested in how companies are responding to climate change (Wasiuzzaman and Subramaniam, 2023). However, not only investors are interested in ESG reporting, but ESG can also be an effective communication tool between a company and its stakeholders.
Considering the interest in ESG performance, several studies have tried to identify the elements that influence organizations' ESG engagement (Abdi et al., 2022; Huang, 2021; Wong et al., 2021). Even though this research showed that financial performance has an impact on ESG performance, it is still unclear how institutional quality drives ESG performance. According to Al-Mamun and Seamer (2022), Institutional quality refers to external influences pushed by the country that encourage firms to take on more social responsibility, as well as economic and financial development, human capital formation and exposure to international trade. El Khoury et al. (2021) suggested future research should concentrate on how the institutional environment influences ESG performance in emerging economies, as there is a concentration of research in developed countries. According to Latapí Agudelo et al. (2020), it is still unclear how national forces affect ESG performance in the energy industry. According to Lee (2021), energy companies have operated directly with natural resources, so they must adhere to regulations, social expectations and international agreements. Despite the fact that the energy industry is deemed environmentally sensitive due to its environmental impact, most previous research on ESG performance has not been conducted in the energy industry (Lee, 2021). For example, previous studies (Buallay, 2019; Shakil et al., 2019) analyzed ESG performance in the banking industry.
Furthermore, Latapí Agudelo et al. (2020) found that the most studies on ESG performance were focused on European and Asian countries, and most of these studies used the Stakeholder Theory as a theoretical lens. As a result, future research should examine ESG performance in the light of other hypotheses (Linnenluecke, 2022). These studies cannot be compared since they used different metrics to quantify ESG performance. As a result, Daugaard and Ding (2022) advocated for new research that would employ a consistent methodology to assess ESG in various institutional contexts. As a result, our research addresses the following question: How do institutional factors influence organizations' ESG performance? In this respect, our study analyzed the role of a country's institutional quality on the ESG performance of its enterprises to fill these gaps in the literature. To that purpose, we examined the ESG performance of 412 organizations based in 19 countries over a four-year period (2016–2019). ESG performance was our dependent variable, and the independent variables were rule of law, economic freedom, education index and international trade freedom, which represented the institutional quality of countries in our investigation.
This study responded to Shin et al.'s (2023) appeal for studies that would include macro and micro level elements. Several research examined environmental challenges without including social and governance practices (Boulagouas et al., 2020; Hristov et al., 2022; Wamane, 2023). In addition, we also addressed the social and governance pillars. ESG performance was rarely studied among enterprises based in emerging economies, according to Sharma et al. (2020). As a result, our analysis included a global sample of enterprises with headquarters in developed and emerging markets. Furthermore, our research assessed institutional quality using variables that have received little attention in the literature.
These findings add to the literature in three ways. First, we corroborated the central thesis of institutional theory (IT), which claims that organizational behavior is influenced by the country's institutional setting. Additionally, we broadened the discussion and addressed the scarcity of research that links ESG performance to institutional quality in the energy industry. Second, our study has management implications. Managers should be aware that varied institutional factors will favor ESG funding. As a result, it is not only a company's internal elements that impact its ESG performance. Third, our findings can help governments improve institutional quality if they want their enterprises to perform better in terms of ESG. For example, by extending the economic freedom of their markets, governments push companies to adopt more ESG practices because there are more stakeholders demanding their corporate performance.
2. Theoretical framework and hypothesis development
IT appears on the seminal studies of DiMaggio and Powell (1983), Meyer and Rowan (1977) and Zucker (1987) and is still regarded as one of the most important theories in organizational studies currently. This theory, which aims to describe how organizations perceive and respond to social change and institutional forces, is complementary to Legitimacy Theory (Bouilloud et al., 2020). As a result, IT focuses on the examination of external factors influencing organizational behavior. Companies are thus viewed as an open system whose quality is impacted by national institutions. In this sense, national institutions are an important concept developed by IT.
Institutions, according to North (1991), are the rulers of the game that control social interactions between people and organizations. Institutions, on the other hand, are defined by Scott (2008) as regulatory, normative and cultural-cognitive aspects that offer stability to social life. By connecting these ideas, institutions are macroeconomic aspects that shape the interaction between companies and their stakeholders. As a result of institutional differences between countries, different organizational arrangements may be developed. These institutional inequalities were identified by DiMaggio and Powell (1983) as pressures that institutions put on businesses. According to these scholars, coercive, normative and mimetic factors compel businesses to behave similarly to survive in the marketplace.
IT demonstrates that enterprises' failure to react to these national pressures may result in legal and societal consequences. As a result, companies within a national system adapt to national institutions to gain legitimacy for their corporate acts.
According to IT, the qualities of a country's institutional environment influence not only a company's financial performance, but also its social and environmental behavior. A substantial number of publications use IT to describe how the institutional environment drives corporate environmental behavior. Different approaches have been used to measure this institutional environment: DiMaggio and Powell's (1983) institutional pressures; Hofstede's (1983) cultural system of countries; Scott's (1995) institutional pillars; Hall and Soskice's (2001) Variety of Capitalism approach; and Whitley's (2003) national business system. The ability of governments to foster economic and social growth is referred to as institutional quality. Governments in nations with higher institutional quality adopt rules and regulations to improve private areas, safeguard property rights and ensure fairness and equality in institutions. Countries with low institutional quality, on the other hand, have weak environmental legislation and institutions that are not transparent (Ashraf et al. 2022a, b).
Previous research has measured institutional quality in a variety of ways. Sun et al. (2019) included five indicators to assess institutional quality: size of government, legal system and property rights, sound money, international trade freedom and credit, labor and business regulation. The level of corruption in the country was used by Zakaria and Bibi (2019) to assess institutional quality. As a result, countries with the lowest levels of corruption have the highest level of institutional quality.
Adusei and Sarpong-Danquah (2021) defined institutional quality as the quality of a country's public governance as represented by six indicators: political stability and absence of violence, corruption control, rule of law, regulatory quality, government effectiveness and voice and accountability. According to Haldar and Sethi (2021), institutional quality represents the efficiency of a country's rules and regulations and can be measured by a country's economic freedom index. According to Ajide and Osinubi (2022), institutional quality is defined as a society's culture and norms. These authors defined institutional quality using four indicators: corruption control, governance stability, bureaucratic quality and law and order. Al-Mamun and Seamer's (2022) investigation proposed that institutional quality determines corporate social responsibility (CSR). According to these scholars, institutional quality may be judged by four factors: rule of law, economic and financial development, human capital formation and exposure to international trade. We chose four variables to represent what Al-Mamun and Seamer (2022) define as institutional quality. Our conceptual model is depicted in Figure 1.
As it can be seen, a corresponding variable was chosen for each pillar of institutional quality. Thus, four research hypotheses are discussed below.
IT backed up the study hypotheses. The rules and norms of the countries, according to this concept, are dominating guidelines for the interaction and behavior of organizations. This theory examines how institutional characteristics of countries shape and adapt organizational challenges over time. Institutional theory was appropriate for our study since it provides a theoretical framework for describing company behavior in various institutional settings (Rahi et al., 2023).
As a result, institutional quality plays a key role in addressing environmental, socioeconomic and governance issues. This is due to regulatory pressure from national institutions on companies to implement measures such as lowering carbon emissions and legislation to define the structure of the board of directors. The investigation of the energy industry is important since it is crucial in many production and consumer activities and economic progress of countries can lead to increased energy consumption. In this aspect, if there are flaws in institutional quality, corporations may ignore negative externalities, resulting in lower ESG performance.
2.1 Hypothesis development
2.1.1 Rule of law
The extent of effective stakeholder protection and the degree to which people trust the country's institutions are indicators of rule of law (Forgione et al., 2020). According to Coluccia et al. (2018), governments in countries with better rule of law are transparent and provide conditions to foster business ethics. Confirming this fact, the study by Pinheiro et al. (2022) showed that companies reflect the institutional context in which they operate; that is, companies have better ESG performance in nations where citizens trust national institutions more. As stakeholders put more pressure on corporations, a better institutional environment promotes companies to embrace ESG practices (El Khoury et al., 2021). Several global studies (Dau et al., 2021; Hamed et al., 2022; Miniaoui et al., 2019) have additionally shown that companies are more committed to ESG issues in countries with higher regulatory quality. Therefore, we argue that:
In countries with greater rule of law, companies have better ESG performance.
2.1.2 Economic development
According to Rosati and Faria (2019), in more open markets, that is, with greater economic freedom, companies seek innovation more frequently, since there are fewer limitations to the behavior of economic players. Additionally, in freer markets, competition between industries is greater, which generates companies' interest in environmental differentiation and transparency (Graafland, 2019; Graafland and Noorderhaven, 2020). Some previous studies (Cai et al., 2019; Hartmann and Uhlenbruck, 2015) have found a positive impact of economic freedom on the social and environmental disclosure of companies. According to the findings of these studies, companies are more likely to behave ethically in countries with greater economic freedom, since in economies that have more open markets, companies have a better relationship with their stakeholders, which facilitates implementation of ESG practices. Therefore, we argue that:
In countries with greater economic freedom, companies have better ESG performance.
2.1.3 Human capital formation
Companies in better-educated countries have more competent human capital, which brings environmental challenges inside the company (Ortas et al., 2019). According to Ioannou and Serafeim (2012), the country's education system has a beneficial impact on ESG performance. These authors argued that in nations where individuals have more access to information, they are more likely to incorporate environmental concerns into their purchase decisions and value enterprises with a better social reputation (Baldini et al., 2018). Furthermore, earlier research (Barkemeyer et al., 2019; Ortas et al., 2019) concluded that where intellectual capital is abundant, stakeholders are more likely to press corporations for improved ESG performance. Therefore, we argue that:
In countries with a higher level of education, companies have better ESG performance.
2.1.4 Exposure to international trade
Globalization and growing international integration have enhanced trade between countries, favoring market access (Tashman et al., 2019). Companies tend to have more stakeholders in nations with more open markets, such as foreign customers, governments and media from other countries (Newman et al., 2018). Furthermore, enterprises that trade on the international market must adhere to environmental regulations not only in their home country, but also abroad. According to Graafland (2019), enterprises in nations with higher exposure to international trade face less government regulation, which they compensate for with improved ESG performance (Acabado et al., 2020; Jackson and Apostolakou, 2010). Therefore, we argue that:
In countries with greater freedom to trade internationally, companies have better ESG performance.
3. Methods
Our population was equivalent to all enterprises in the energy industry of the G20 countries (an association comprised of the world's 19 most developed countries plus the European Union). We limited our sample to companies in the energy industry that had ESG, financial and governance data in the Refinitiv Eikon database (previously known as Thomson Reuters Eikon). We investigated the energy industry since companies in this sector use natural resources directly in their operations and are responsible for a considerable portion of carbon emissions into the environment (Papadis and Tsatsaronis, 2020). All the companies in the sample were publicly traded on the stock exchange.
Our study included 412 enterprises from the G20's 19 member countries and spanned four years, from 2016 to 2019. According to Orzes et al. (2020), after signing the UN Global Compact, companies raised their interest in ESG challenges. We avoided the years 2020 and 2021 since they were the years of the Covid-19 pandemic, which could skew the results. Companies lowered their investments in ESG during the pandemic period, according to Cardillo et al. (2023) due to their concerns with their financial survival. Table 1 presents the sample.
Number of companies by country
| Country | No of companies | Percentage | No of observations by company |
|---|---|---|---|
| Argentina | 4 | 0.97 | 16 |
| Australia | 25 | 6.07 | 100 |
| Brazil | 7 | 1.70 | 28 |
| Canada | 65 | 15.78 | 260 |
| China | 33 | 8.01 | 132 |
| France | 10 | 2.43 | 40 |
| Germany | 8 | 1.94 | 32 |
| India | 9 | 2.18 | 36 |
| Indonesia | 6 | 1.46 | 24 |
| Italy | 5 | 1.21 | 20 |
| Japan | 7 | 1.70 | 28 |
| Korea | 6 | 1.46 | 24 |
| Mexico | 2 | 0.49 | 8 |
| Russia | 10 | 2.43 | 40 |
| Saudi Arabia | 2 | 0.49 | 8 |
| South Africa | 2 | 0.49 | 8 |
| Turkey | 3 | 0.73 | 12 |
| The United Kingdom | 24 | 5.83 | 96 |
| The United States | 184 | 44.66 | 736 |
| Total | 412 | 100.00 | 1,648 |
| Country | No of companies | Percentage | No of observations by company |
|---|---|---|---|
| Argentina | 4 | 0.97 | 16 |
| Australia | 25 | 6.07 | 100 |
| Brazil | 7 | 1.70 | 28 |
| Canada | 65 | 15.78 | 260 |
| China | 33 | 8.01 | 132 |
| France | 10 | 2.43 | 40 |
| Germany | 8 | 1.94 | 32 |
| India | 9 | 2.18 | 36 |
| Indonesia | 6 | 1.46 | 24 |
| Italy | 5 | 1.21 | 20 |
| Japan | 7 | 1.70 | 28 |
| Korea | 6 | 1.46 | 24 |
| Mexico | 2 | 0.49 | 8 |
| Russia | 10 | 2.43 | 40 |
| Saudi Arabia | 2 | 0.49 | 8 |
| South Africa | 2 | 0.49 | 8 |
| Turkey | 3 | 0.73 | 12 |
| The United Kingdom | 24 | 5.83 | 96 |
| The United States | 184 | 44.66 | 736 |
| Total | 412 | 100.00 | 1,648 |
Source(s): Table created by author
As it can be seen, the final sample of 412 international enterprises was based in 19 countries and represented a total of 1,648 observations (Observations were made using the following formula: number of companies per country x number of years analyzed). The country with the most representation was the United States, which had 184 enterprises, accounting for 44.66% of the total. Following that, Canada and China had the most companies, with 65 and 33, respectively. On the other hand, Mexico, Saudi Arabia and South Africa had only 2 companies, which meant the smallest representation in our sample. These three countries made up less than 2% of the sample.
According to Ortas et al. (2019), ESG performance is a multidimensional construct formed of social and environmental consequences of companies related to various stakeholders. This variable is continuous and has a range of 0–100 (businesses with the lowest ESG Performance). This variable was separated into three pillars: ESG. Each of these pillars was likewise awarded a score ranging from 0 to 100. Table 2 lists the variables, measurements and data sources.
Variable definitions, measurements, and data sources
| Variables | Definition | Source |
|---|---|---|
| ESG | ESG Performance: continuous variable ranging from 0 (lowest corporate ESG performance) to 100 (highest corporate ESG performance) | Refinitiv Eikon database |
| ENVIR | Environmental performance: continuous variable that varies from 0 (lowest environmental performance) to 100 (highest environmental performance) | Refinitiv Eikon database |
| SOCIA | Social performance: continuous variable that varies from 0 (lowest social performance) to 100 (highest social performance) | Refinitiv Eikon database |
| GOVER | Governance performance: continuous variable that ranges from 0 (lowest governance performance) to 100 (higher governance performance) | Refinitiv Eikon database |
| RULLAW | Rule of law: continuous variable that captures the degree to which agents trust and comply with society's rules, ranging from −2.5 (lowest trust in society's rules) to +2.5 (highest trust) | Worldwide Governance Indicators, World Bank |
| ECOFRE | Economic Freedom: continuous variable that captures government size, regulatory efficiency, and market openness, ranging from 0 (least freedom) to 100 (most freedom) | Heritage Foundation |
| EDUIND | Education index: continuous variable that measures the average years of schooling (adults) and expected years of schooling (children), ranging from 11.90 (lowest educational level) to 23.10 (highest educational level) | United Nations Development Programme |
| TRAINT | Freedom to trade internationally: continuous variable that measures the ease of companies to export their products to other countries, ranging from 3.99 (least freedom to trade) to 10 (greater freedom) | Fraser Institute |
| MARKCAP | Market Capitalization: refers to the total dollar market value of a company's outstanding shares | Refinitiv Eikon database |
| ROE | Return on Equity: It is the ratio of net income to equity | Refinitiv Eikon database |
| FIRMSIZE | Firm Size: Natural logarithm of companies' total assets | Refinitiv Eikon database |
| LEVERAGE | Leverage: firms' long-term debt divided by common equity | Refinitiv Eikon database |
| BOARDSIZE | Board size: Number of directors on board | Refinitiv Eikon database |
| BOARDIND | Board independence: Total number of independent directors on boards/total number of directors on boards | Refinitiv Eikon database |
| BOARDIVER | Board diversity: Percentage of women on the board of directors | Refinitiv Eikon database |
| CSRCOM | Corporate Social Responsibility Committee: Dummy variable: 1 = if the company has a CSR committee, and 0 = otherwise | Refinitiv Eikon database |
| UNSIGN | UN Global Compact Signatory: Dummy variable: 1 = if the company is a signatory of the UN Global Compact, and 0 = otherwise | Refinitiv Eikon database |
| Variables | Definition | Source |
|---|---|---|
| ESG | ESG Performance: continuous variable ranging from 0 (lowest corporate ESG performance) to 100 (highest corporate ESG performance) | Refinitiv Eikon database |
| ENVIR | Environmental performance: continuous variable that varies from 0 (lowest environmental performance) to 100 (highest environmental performance) | Refinitiv Eikon database |
| SOCIA | Social performance: continuous variable that varies from 0 (lowest social performance) to 100 (highest social performance) | Refinitiv Eikon database |
| GOVER | Governance performance: continuous variable that ranges from 0 (lowest governance performance) to 100 (higher governance performance) | Refinitiv Eikon database |
| RULLAW | Rule of law: continuous variable that captures the degree to which agents trust and comply with society's rules, ranging from −2.5 (lowest trust in society's rules) to +2.5 (highest trust) | Worldwide Governance Indicators, World Bank |
| ECOFRE | Economic Freedom: continuous variable that captures government size, regulatory efficiency, and market openness, ranging from 0 (least freedom) to 100 (most freedom) | Heritage Foundation |
| EDUIND | Education index: continuous variable that measures the average years of schooling (adults) and expected years of schooling (children), ranging from 11.90 (lowest educational level) to 23.10 (highest educational level) | United Nations Development Programme |
| TRAINT | Freedom to trade internationally: continuous variable that measures the ease of companies to export their products to other countries, ranging from 3.99 (least freedom to trade) to 10 (greater freedom) | Fraser Institute |
| MARKCAP | Market Capitalization: refers to the total dollar market value of a company's outstanding shares | Refinitiv Eikon database |
| ROE | Return on Equity: It is the ratio of net income to equity | Refinitiv Eikon database |
| FIRMSIZE | Firm Size: Natural logarithm of companies' total assets | Refinitiv Eikon database |
| LEVERAGE | Leverage: firms' long-term debt divided by common equity | Refinitiv Eikon database |
| BOARDSIZE | Board size: Number of directors on board | Refinitiv Eikon database |
| BOARDIND | Board independence: Total number of independent directors on boards/total number of directors on boards | Refinitiv Eikon database |
| BOARDIVER | Board diversity: Percentage of women on the board of directors | Refinitiv Eikon database |
| CSRCOM | Corporate Social Responsibility Committee: Dummy variable: 1 = if the company has a CSR committee, and 0 = otherwise | Refinitiv Eikon database |
| UNSIGN | UN Global Compact Signatory: Dummy variable: 1 = if the company is a signatory of the UN Global Compact, and 0 = otherwise | Refinitiv Eikon database |
Source(s): Table created by author
The independent variables represented the countries' institutional quality. According to Al-Mamun and Seamer (2022), four country characteristics can be used to assess institutional quality: rule of law, economic and financial growth, human capital formation and exposure to international trade. We selected variables to represent each of these qualities. Rule of law, economic freedom, education index and freedom to trade internationally are continuous variables and were collected from reports of international bodies such as the World Bank, Heritage Foundation, United Nations Development Program and Fraser Institute.
In econometric models, the impact of institutional quality on ESG performance is controlled by using financial and governance variables to prevent estimate bias. These factors were chosen because they have been traditionally related to a company's ESG level. Market capitalization, return on equity, business size and leverage are the financial control variables. Board size, board independence, board diversity, CSR committee and UN Global Compact Signatory are the governance control variables. Previous research has demonstrated that these variables can affect ESG performance (Atan et al., 2018; Buallay, 2019; Kumar et al., 2022; Shakil et al., 2019).
Panel data analysis was suitable for answering our study question because it captures ESG variation across time. Panel data approaches can identify impacts that are not visible in cross-sectional or time-series data (Hair et al., 2019). Individual heterogeneity is controlled for using fixed or random effects models. In this way, they accounted for unobservable aspects of the companies under consideration. Under rigorous conditions of the exogeneity of the regressors, panel data models yield unbiased statistical estimates (Ferreira et al., 2022). As we had a database of several years, the data panel was suitable because it captured the effect of years on ESG performance.
To test our hypotheses, we ran the following econometric models:
where the subscript “i” refers to the company, “t” represents the time, β is the estimated parameter, θ refers to the unobservable effects which are time-variant and specific of each company, and ɛ indicates the error. Models were estimated using panel data regression with fixed and random effects.
Company-specific effects are associated with the explanatory variables in the fixed effects model (Hair et al., 2019). The Hausman test identifies the optimum model between fixed and random effects. The Hausman test's null and alternative hypotheses are that the random and fixed effects are appropriate, respectively. In our case, the preferred model is the fixed effects model, since the panel data with random effects was rejected by the test of Hausman (Hair et al., 2019).
We operationalized tests to quantify the data's multicollinearity, heteroscedasticity and endogeneity to validate and raise confidence in our conclusions. The Variance Inflation Factor (VIF) was used to assess multicollinearity. VIF less than 10 indicates the lack of multicollinearity. The Breusch-Pagan test was used to determine heteroscedasticity. The Durbin-Wu Hausman test (Ferreira et al., 2022) was used to control endogeneity issues. Additionally, we ran robustness tests, excluding US enterprises because they made up a sizable portion of our sample.
4. Results and discussion
4.1 Descriptive statistics
Table 3 presents a summary of the descriptive statistics for all variables examined. The average of the dependent variable (ESG performance) was 41.43. In terms of the independent variables, the rule of law average was 1.25, indicating more trust in society's laws. The economic freedom index average was 72.35, indicating that it was closer to the element of more freedom, but the education index average was 16.35, indicating that, in general, the companies in the sample were based in nations with higher levels of education. Finally, the freedom of international trade average was 7.70, indicating that it was easier for companies to promote their products in other nations.
Descriptive analysis
| Variable | Obs | Mean | Std. Dev | Min | Max |
|---|---|---|---|---|---|
| ESG | 1,437 | 41,439 | 21,274 | 0.31 | 91.48 |
| ENVIR | 1,437 | 35,265 | 27,169 | 0.00 | 96.28 |
| SOCIA | 1,437 | 41,417 | 24,213 | 0.66 | 94.97 |
| GOVER | 1,437 | 50,377 | 23,664 | 0.16 | 98.56 |
| RULLAW | 1,437 | 1,254 | 0.763 | −0.79 | 1.84 |
| ECOFRE | 1,437 | 72,359 | 8,407 | 43.8 | 81.00 |
| EDUIND | 1,437 | 16,351 | 1,939 | 11.9 | 23.10 |
| TRAINT | 1,437 | 7,702 | 0.612 | 5.20 | 8.77 |
| MARKCAP | 1,437 | 9,162 | 0.912 | 5.61 | 12.28 |
| ROE | 1,437 | 0.024 | 0.737 | −20.09 | 2.64 |
| FIRMSIZE | 1,437 | 9,490 | 0.869 | 6.54 | 11.76 |
| LEVERAGE | 1,437 | 0.763 | 2,152 | −1.51 | 31.4 |
| BOARDSIZE | 1,437 | 8,721 | 2,758 | 1.00 | 22.0 |
| BOARDIND | 1,437 | 65,089 | 22,974 | 0.00 | 100 |
| BOARDIVER | 1,437 | 15,237 | 12,446 | 0.00 | 62.5 |
| CSRCOM | 1,437 | 0.582 | 0.493 | 0.00 | 1.00 |
| UNSIGN | 1,437 | 0.128 | 0.335 | 0.00 | 1.00 |
| Variable | Obs | Mean | Std. Dev | Min | Max |
|---|---|---|---|---|---|
| ESG | 1,437 | 41,439 | 21,274 | 0.31 | 91.48 |
| ENVIR | 1,437 | 35,265 | 27,169 | 0.00 | 96.28 |
| SOCIA | 1,437 | 41,417 | 24,213 | 0.66 | 94.97 |
| GOVER | 1,437 | 50,377 | 23,664 | 0.16 | 98.56 |
| RULLAW | 1,437 | 1,254 | 0.763 | −0.79 | 1.84 |
| ECOFRE | 1,437 | 72,359 | 8,407 | 43.8 | 81.00 |
| EDUIND | 1,437 | 16,351 | 1,939 | 11.9 | 23.10 |
| TRAINT | 1,437 | 7,702 | 0.612 | 5.20 | 8.77 |
| MARKCAP | 1,437 | 9,162 | 0.912 | 5.61 | 12.28 |
| ROE | 1,437 | 0.024 | 0.737 | −20.09 | 2.64 |
| FIRMSIZE | 1,437 | 9,490 | 0.869 | 6.54 | 11.76 |
| LEVERAGE | 1,437 | 0.763 | 2,152 | −1.51 | 31.4 |
| BOARDSIZE | 1,437 | 8,721 | 2,758 | 1.00 | 22.0 |
| BOARDIND | 1,437 | 65,089 | 22,974 | 0.00 | 100 |
| BOARDIVER | 1,437 | 15,237 | 12,446 | 0.00 | 62.5 |
| CSRCOM | 1,437 | 0.582 | 0.493 | 0.00 | 1.00 |
| UNSIGN | 1,437 | 0.128 | 0.335 | 0.00 | 1.00 |
Source(s): Table created by author
Furthermore, the market capitalization average was 9.16, the return on equity average was 0.02, the company's total assets average was 9.49, the leverage average was 0.76, the board size average was 8.72, and board independence average was 65.08. The average percentage of women on the board of directors was 15.23%, showing that the companies in the sample had low female participation on the boards. The corporate social responsibility committee showed an average of 0.58, indicating that most companies had a sustainability committee. Finally, the average of the UN Global Compact adoption was 0.12, which revealed that most companies in the sample had not signed such global agreement. The findings revealed significant variances in standard deviation amongst the companies evaluated in terms of ESG and their pillars. This could imply that ESG performance varied depending on the country's institutional environment. Companies may have faced a range of institutional constraints and, as a result, tended to perform better in terms of ESG in specific circumstances.
4.2 Multivariate analysis and discussion
Table 4 summarizes the results of the four models developed to assess the effects of independent variables on ESG performance and their variations (environmental, social and corporate governance dimensions).
Multivariate analysis results
| Variable | Model I– ESG | Model II– E | Model III– S | Model IV– G |
|---|---|---|---|---|
| Coef. (sig) | Coef. (sig) | Coef. (sig) | Coef. (sig) | |
| RULLAW | −11.581*** | −7.479 (0.002) | −9.731*** | −19.57*** |
| ECOFRE | 0.371** | 0.313 (0.186) | 0.302 (0.198) | 0.618*** |
| EDUIND | −0.216 (0.490) | −1.074 (0.210) | −0.344 (0.403) | 0.995** |
| TRAINT | 6.764*** | 5.278*** | 7.726*** | 7.671*** |
| MARKCAP | 2.542** | 5.522*** | 7.131*** | −6.494*** |
| ROE | −0.864 (0.557) | −1.741 (0.371) | −0.383 (0.843) | −0.041 (0.984) |
| FIRMSIZE | 5.344*** | 8.240*** | −0.634 (0.729) | 8.780*** |
| LEVERAGE | 0.113 (0.581) | 0.354 (0.191) | 0.064 (0.812) | −0.069 (0.809) |
| BOARDSIZE | 0.177 (0.372) | 0.147 (0.575) | −0.059 (0.821) | 0.505** |
| BOARDIND | 0.059*** | −0.112*** | −0.032 (0.320) | 0.416*** |
| BOARDIVER | 0.324*** | 0.314*** | 0.254*** | 0.426*** |
| CSRCOM | 15.067*** | 15.62*** | 18.29*** | 10.42*** |
| UNSIGN | 10.732*** | 11.72*** | 15.45*** | 1.801 (0.350) |
| _CONS | −113.3*** | −139.1*** | −94.37*** | −112.3*** |
| Obs | 954 | 954 | 954 | 954 |
| R2 | 0.5729 | 0.5455 | 0.4545 | 0.3565 |
| VIF | 3.68 | 3.68 | 3.68 | 3.68 |
| Breusch-Pagan test | 0.326 | 0.592 | 0.2759 | 0.0261 |
| Durbin–Watson test | No endogenous | No endogenous | No endogenous | No endogenous |
| Wald x2 test | 1,274.48 | 1,142.23 | 797.47 | 518.96 |
| Hausman test | Fixed effects | Fixed effects | Fixed effects | Fixed effects |
| Variable | ||||
|---|---|---|---|---|
| Coef. (sig) | Coef. (sig) | Coef. (sig) | Coef. (sig) | |
| RULLAW | −11.581*** | −7.479 (0.002) | −9.731*** | −19.57*** |
| ECOFRE | 0.371** | 0.313 (0.186) | 0.302 (0.198) | 0.618*** |
| EDUIND | −0.216 (0.490) | −1.074 (0.210) | −0.344 (0.403) | 0.995** |
| TRAINT | 6.764*** | 5.278*** | 7.726*** | 7.671*** |
| MARKCAP | 2.542** | 5.522*** | 7.131*** | −6.494*** |
| ROE | −0.864 (0.557) | −1.741 (0.371) | −0.383 (0.843) | −0.041 (0.984) |
| FIRMSIZE | 5.344*** | 8.240*** | −0.634 (0.729) | 8.780*** |
| LEVERAGE | 0.113 (0.581) | 0.354 (0.191) | 0.064 (0.812) | −0.069 (0.809) |
| BOARDSIZE | 0.177 (0.372) | 0.147 (0.575) | −0.059 (0.821) | 0.505** |
| BOARDIND | 0.059*** | −0.112*** | −0.032 (0.320) | 0.416*** |
| BOARDIVER | 0.324*** | 0.314*** | 0.254*** | 0.426*** |
| CSRCOM | 15.067*** | 15.62*** | 18.29*** | 10.42*** |
| UNSIGN | 10.732*** | 11.72*** | 15.45*** | 1.801 (0.350) |
| _CONS | −113.3*** | −139.1*** | −94.37*** | −112.3*** |
| Obs | 954 | 954 | 954 | 954 |
| R2 | 0.5729 | 0.5455 | 0.4545 | 0.3565 |
| VIF | 3.68 | 3.68 | 3.68 | 3.68 |
| Breusch-Pagan test | 0.326 | 0.592 | 0.2759 | 0.0261 |
| Durbin–Watson test | No endogenous | No endogenous | No endogenous | No endogenous |
| Wald x2 test | 1,274.48 | 1,142.23 | 797.47 | 518.96 |
| Hausman test | Fixed effects | Fixed effects | Fixed effects | Fixed effects |
Note(s): ***p < 0.01. **p < 0.05. *p < 0.10
Source(s): Table created by author
Initially, the rule of law had a negative effect on the dependent variable based on the results of the independent factors, opposing hypothesis 1. This hypothesis assumed that companies perform better in countries where the rule of law is stronger. The findings indicated that the more agents believed society's laws, the less engaged they were with ESG performance.
The hypothesis 1 result contrasted the findings of Pinheiro et al. (2022). In spite of this, we may speculate that in countries where people trusted the rule of law more, companies performed worse in terms of ESG as they were not required to disclose such information, for example, in a formal report. In the absence of a formal environmental report, society in these countries can rely on company performance and expect them to act dependably.
The findings of economic freedom support hypothesis 2, namely that enterprises perform better in countries with higher economic freedom. Greater economic independence in the country had a positive impact on governance performance, as evidenced by the data. This finding is consistent with earlier research (Cai et al., 2019; Hartmann and Uhlenbruck, 2015). Greater economic freedom fosters and facilitates the establishment of new companies, resulting in increased competition. In many cases, ESG performance can provide a competitive advantage, implying that the company not only produces a high-quality product but also behaves responsibly.
Given the competitiveness, stakeholders in more economically free countries are expected to put more pressure on companies to be committed to responsible behavior, driving them to enhance ESG performance. Companies based in liberal market economies, according to Jackson and Apostolakou (2010) and Mooneeapen et al. (2022), tend to have more voluntary disclosure of environmental and social issues to compensate for the lack of government control.
The degree of education had a significant and positive impact, showing that a higher level of education resulted in a higher level of governance. Companies in countries with higher levels of education had better corporate governance. This finding was supported by Ortas et al. (2019). The education indicator, on the other hand, showed no noticeable effect on ESG performance. As a result, hypothesis 3 was partially refuted. ESG activities, according to Ashraf et al. (2022a, b), serve as a supplement in countries with a prominent level of education. Higher education can help people improve their critical thinking skills and facilitate discussions that extend beyond traditional financial issues.
When a country has more intellectual capital available, its corporations are more likely to have greater corporate governance because people in the country are thought to have better access to education to operate businesses correctly. Using Ioannou and Serafeim's (2012) research, we may conclude that individuals in better educated countries have greater access to information and are more aware of the company's responsibilities. In this situation, companies concentrate their attention on new challenges (ESG performance) in addition to traditional financial considerations.
The fourth hypothesis, claiming that countries with greater international trade flexibility outperform others, was confirmed. According to the findings, increased international trade flexibility enabled companies to perform better in terms of environmental, social and corporate governance. Previous study has yielded comparable results (Acabado et al., 2020; Hartmann and Uhlenbruck, 2015; Wang et al., 2023).
Private market competitiveness is higher in places where trading is easier (Graafland and Noorderhaven, 2020). Competitiveness promotes environmental and social advances. As a result, when organizations deal with more stakeholders and a foreign market, they must compete harder. Furthermore, economic freedom has the potential to mitigate the effects of corruption and urge enterprises to take greater responsibility for their impacts on community well-being (Daugaard and Ding, 2022).
Analyzing the control variables revealed that the higher the company's overall market value, the better its ESG performance. According to the data, the larger the company, the better its ESG performance. In practice, larger organizations have more financial resources to comply with ESG policies. These findings are consistent with recent research (Orzes et al., 2020; Tashman et al., 2019; Wong et al., 2021) that has demonstrated that larger organizations have a greater impact on society, which raises their interest in ESG performance.
The long-term debt of the companies, or leverage, had no meaningful effect on the dependent variables. This suggests that the impacts of corporate leverage on ESG performance are not yet obvious. The findings showed that a larger board of directors improves a company's corporate governance performance. As people come from various backgrounds, a larger board size fosters diversity of ideas. As a result, larger boards can favor corporate governance excellence.
A higher ratio of women on the board of directors results in better ESG and variation performance, demonstrating that having more women on the board of directors helps companies improve their ESG performance. This confirms the findings of Issa and Hanaysha (2023). Women on boards are twice as likely to have a doctorate than men (Hillman and Dalziel, 2003). Women on boards are also more likely to have additional technical knowledge unrelated to management, allowing them to bring agenda items such as social responsibility to meetings (Gaio and Gonçalves, 2022; Gurol and Lagasio, 2023).
In terms of the social responsibility committee, it was found that when a company has one, it performs better in terms of ESG and its pillars. Additionally, the company's participation in the UN Global Compact has a beneficial impact on ESG performance. The results allowed us to conclude that the presence of a sustainability committee and the adoption of the Global Compact are crucial for companies to have greater ESG performance, as shown in previous studies (Abdi et al., 2022; Latapí Agudelo et al., 2020; Orzes et al., 2020; Rosati and Faria, 2019).
In addition, our findings allowed us to identify institutional elements that influence company performance in respect to ESG challenges in both emerging and developed countries. Companies based in the same organizational field (country) exhibit comparable behavior when it comes to disclosing their ESG policies. The quality of local institutions might generate disparities in ESG practices across corporations in different nations. Companies are not under enough pressure to address ESG issues in contexts with low institutional quality.
Although our study demonstrated that ESG performance is affected by a country's macroeconomic situations, it also found that a stronger rule of law is not required for corporations to engage with ESG. The evidence suggests that companies, even in government structures without transparency, can act on environmental and social issues. This is in line with the findings of the Ashraf et al. (2022a, b) study, which found that corporations located in countries with low institutional quality might use ESG performance as a strategy to compensate for the country's social difficulties.
4.3 Robustness analysis
Table 5 shows the findings for the correlations of influence of the independent variables on the dependent variables (ESG and its variations), excluding companies headquartered in the United States. We conducted this robustness analysis to confirm the prior results, as the United States has a large number of enterprises, which could skew the research results.
Robustness analysis results
| Variable | Model I– ESG | Model II– E | Model III– S | Model IV– G |
|---|---|---|---|---|
| Coef. (sig) | Coef. (sig) | Coef. (sig) | Coef. (sig) | |
| RULLAW | −8.632*** | −4.158* | −7.742*** | −15.49*** |
| ECOFRE | 0.284 (0.120) | 0.179 (0.454) | 0.187 (0.459) | 0.615*** |
| EDUIND | −0.704 (0.127) | −1.647 (0.178) | −0.615 (0.161) | 0.308 (0.508) |
| TRAINT | 7.671*** | 6.602*** | 8.369*** | 8.433*** |
| MARKCAP | 4.102** | 7.592*** | 7.449*** | −3.806 (0.141) |
| ROE | −2.435 (0.265) | −3.561 (0.213) | −1.553 (0.607) | −1.304 (0.684) |
| FIRMSIZE | 3.738** | 5.932** | −2.339 (0.362) | 8.173*** |
| LEVERAGE | 0.117 (0.705) | 0.306 (0.449) | −0.008 (0.985) | 0.216 (0.633) |
| BOARDSIZE | −0.053 (0.823) | −0.137 (0.657) | −0.268 (0.414) | 0.264 (0.447) |
| BOARDIND | 0.048** | −0.104*** | −0.053 (0.198) | 0.391*** |
| BOARDIVER | 0.204*** | 0.160** | 0.162** | 0.295*** |
| CSRCOM | 13.34*** | 12.84*** | 17.81*** | 8.237*** |
| UNSIGN | 9,355*** | 10.12*** | 15.12*** | −0.421 (0.850) |
| _CONS | −99.70*** | −121.2*** | −68.77*** | −119.6*** |
| Obs | 566 | 566 | 566 | 566 |
| R2 | 0.5234 | 0.5069 | 0.4019 | 0.3191 |
| VIF | 3.92 | 3.92 | 3.92 | 3.92 |
| Breusch-Pagan test | 0.0005 | 0.0148 | 0.0725 | 0.0007 |
| Durbin–Watson test | No endogenous | No endogenous | No endogenous | No endogenous |
| Wald x2 test | 608.01 | 567.93 | 374.22 | 258.48 |
| Hausman test | Fixed effects | Fixed effects | Fixed effects | Fixed effects |
| Variable | ||||
|---|---|---|---|---|
| Coef. (sig) | Coef. (sig) | Coef. (sig) | Coef. (sig) | |
| RULLAW | −8.632*** | −4.158* | −7.742*** | −15.49*** |
| ECOFRE | 0.284 (0.120) | 0.179 (0.454) | 0.187 (0.459) | 0.615*** |
| EDUIND | −0.704 (0.127) | −1.647 (0.178) | −0.615 (0.161) | 0.308 (0.508) |
| TRAINT | 7.671*** | 6.602*** | 8.369*** | 8.433*** |
| MARKCAP | 4.102** | 7.592*** | 7.449*** | −3.806 (0.141) |
| ROE | −2.435 (0.265) | −3.561 (0.213) | −1.553 (0.607) | −1.304 (0.684) |
| FIRMSIZE | 3.738** | 5.932** | −2.339 (0.362) | 8.173*** |
| LEVERAGE | 0.117 (0.705) | 0.306 (0.449) | −0.008 (0.985) | 0.216 (0.633) |
| BOARDSIZE | −0.053 (0.823) | −0.137 (0.657) | −0.268 (0.414) | 0.264 (0.447) |
| BOARDIND | 0.048** | −0.104*** | −0.053 (0.198) | 0.391*** |
| BOARDIVER | 0.204*** | 0.160** | 0.162** | 0.295*** |
| CSRCOM | 13.34*** | 12.84*** | 17.81*** | 8.237*** |
| UNSIGN | 9,355*** | 10.12*** | 15.12*** | −0.421 (0.850) |
| _CONS | −99.70*** | −121.2*** | −68.77*** | −119.6*** |
| Obs | 566 | 566 | 566 | 566 |
| R2 | 0.5234 | 0.5069 | 0.4019 | 0.3191 |
| VIF | 3.92 | 3.92 | 3.92 | 3.92 |
| Breusch-Pagan test | 0.0005 | 0.0148 | 0.0725 | 0.0007 |
| Durbin–Watson test | No endogenous | No endogenous | No endogenous | No endogenous |
| Wald x2 test | 608.01 | 567.93 | 374.22 | 258.48 |
| Hausman test | Fixed effects | Fixed effects | Fixed effects | Fixed effects |
Note(s): ***p < 0.01. **p < 0.05. *p < 0.10
Source(s): Table created by author
In general, the robustness analysis results were consistent with earlier findings. The rule of law was detrimental to ESG performance and its pillars. When economic independence is considered, the results show that it has a considerable impact on the corporate governance variable. This suggests that companies perform better in governance in countries with greater economic freedom. The presence of American corporations has influenced economic freedom, which has had a beneficial and considerable impact on ESG performance. The findings confirmed that companies headquartered in countries where it is simpler to market their products abroad do better in terms of ESG.
By analyzing market capitalization and company size variables, we can confirm that larger companies tend to make more investments in ESG issues to meet the expectations of their stakeholders. The presence of more independent directors has a beneficial impact on ESG performance, confirming prior findings. Finally, gender diversity on the board, the inclusion of a CSR committee and the adoption of the Global Compact have a beneficial impact on ESG performance and its variations.
These new findings revealed that, despite the majority of the companies in our sample were based in the United States, this did not impair the robustness of the analysis, but instead, complemented the data interpretations.
4.4 Theoretical and practical implications
Several implications can be derived from this analysis. First, our results showed that not only do internal factors interfere with ESG performance, but also institutional factors have a significant effect on companies' performance in environmental and social issues. These findings reinforced the theoretical foundations of IT, which states that organizational behavior is shaped by the national context in which firms operate. The importance of this study also resides in the fact that there are still no studies that relate institutional quality to ESG performance specifically in the energy industry.
According to institutional theory, organizations act within social structures and rules that affect their decisions. In this paper, we presented new evidence that the performance of companies in the energy industry is shaped by the institutional characteristics of the countries in which they are headquartered. Our evidence suggests that institutional quality is a complement for companies to achieve higher ESG performance.
In addition to the academic level, the results have managerial implications. The findings allow managers to better understand how the institutional environment demands ESG efforts by companies. Companies in the energy sector may have different ESG priorities depending on the country in which they operate. Stakeholders expect companies to meet their needs and by having a higher ESG performance, companies bring benefits not only to shareholders, but also to society. Our results encourage managers to look at institutional quality at the country level. In addition, we suggest that managers implement a CSR committee and adopt the Global Compact if they want their companies to have better ESG performance.
Furthermore, our data may be relevant to international policymakers by reinforcing that greater economic freedom, better education and greater international trade freedom promote higher ESG performance. We also propose that authorities consider establishing legislation to safeguard the integrity and trustworthiness of ESG information presented. Because ESG actions are not required to be published, corporate report organizers choose the information disclosed each year, which can lead to a lack of uniformity and comparability among companies. As a result, regulators must devise procedures to limit managers'/organizers' opportunism in order to increase the objectivity, efficiency and verifiability of ESG data.
4.5 Limitations and directions for future research
There are limitations to this study. The investigation focused exclusively at large corporations in the energy industry, thus the conclusions cannot be applied to small businesses or other industries. Our metric to measure ESG performance is based on indicators from the Thomson Reuters Eikon database. Therefore, new research can create other ESG performance indicators that are less susceptible to greenwashing, that is, the metrics must also measure the quality of the information disclosed by companies.
Further studies should cover a bigger sample of developing nations as well as additional business sectors. Additionally, new studies can understand how the institutional environment moderates the relationship between ESG performance and corporate green innovation. Furthermore, researchers should consider institutional-level elements that have received little attention in the literature. Finally, we recommend collecting ESG performance from different databases.
5. Conclusions
Using institutional variety across nations, this article sought to investigate the influence of institutional quality of countries on their companies' ESG performance. We assessed institutional quality using four variables: rule of law, economic freedom, education index and international trade freedom. We selected indicators from the Refinitiv Eikon database to quantify ESG performance, as such database contains 70 major indicators linked to ESG performance and is one of the most credible sources for comparing data from companies all over the world.
The findings showed that institutional quality is important for ESG performance. Companies headquartered in nations with greater economic freedom and countries that encourage their companies' international trade tend to be more engaged with ESG issues. Furthermore, the findings confirmed that corporations have stronger governance in countries with higher levels of education. Our research shows that, in addition to boosting economic development, as previous studies have demonstrated, institutional quality encourages greater ESG performance.
The authors would like to thank Editor Prof Malin Song for his excellent support during the review process and two anonymous reviewers for their valuable comments. The previous version of this paper was presented at the Seminários em Administração (Semead) Conference in São Paulo. At this conference, this research also benefited greatly from the suggestions made by professors Flavio Hourneaux Junior and Larissa Marchiori Pacheco. The authors are also grateful for the financial support provided by CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil).
Since acceptance of this article, the following author have updated their affiliations: Alan Bandeira Pinheiro is at the NEOMA Business School, Rouen, France.

