International Journal of Innovation and Economic Development

Volume 11, Issue 5, December 2025, Pages 7-20

---

The relationship between ESG criteria and
economic growth: A study on Stoxx Europe 600
company countries

URL: https://doi.org/10.18775/ijied.1849-7551-7020.2015.115.2001
DOI: 10.18775/ijied.1849-7551-7020.2015.115.2001

¹Pınar Yiğitdoğan, ²Halil Tunalı

^İ stanbul University, Institute of Social Sciences, Department of Economics, Istanbul, Turkey

Abstract

This study examines the impact of environmental, social, and governance (ESG) criteria on economic growth within a macroeconomic framework. While labor, physical capital, and human capital are considered key determinants in classical growth models, ESG criteria have become an increasingly important factor in terms of economic stability and productivity in recent years. In this context, the study aims to analyze the relationship between sustainability and growth by integrating ESG indicators into an extended neoclassical growth model of the Solow–Mankiw–Romer–Weil type. The empirical analysis was conducted using annual panel data covering the period 2004–2023 for 16 European countries with companies listed on the STOXX Europe 600 index. ESG scores at the firm level were aggregated at the country level, and model estimates were obtained using Driscoll–Kraay standard errors to address cross-sectional dependence, autocorrelation, and heteroskedasticity issues. The findings show that physical capital accumulation continues to be the key determinant of economic growth, while the ESG indicator has a negative impact on real GDP in the short term. This effect is associated with compliance and regulatory costs arising during the transition to sustainable production and governance structures. The results reveal that the impact of ESG criteria on economic growth varies depending on the time dimension and corporate structure, and that the long-term gains of sustainability policies should be evaluated taking into account short-term costs.

Keywords

ESG criteria; economic growth; sustainability; panel data analysis; European economies

1. Introduction

Identifying the fundamental determinants of economic growth has long been a central theme in macroeconomic literature. Since the seminal work of Solow (1957) and Swan (1956), the neoclassical growth model has provided a robust framework for analyzing the role of capital accumulation and population growth in determining long-term income levels. Subsequently, Mankiw, Romer, and Weil (1992) showed that the standard Solow/Swan model cannot explain the magnitude of income differences between countries without taking human capital accumulation into account.

In the contemporary economic context, environmental, social, and governance (ESG) performance has emerged as a new dimension that can influence the productive capacity of economies and their long-term stability. Alongside labor, physical capital, human capital, and technological progress—traditionally recognized as the main drivers of growth—ESG criteria, initially adopted on a voluntary basis by companies, have gradually become a structural element of economic models following their integration into European regulatory frameworks since the 2010s. This development means that ESG is no longer considered solely as an ethical choice, but as an institutional factor influencing the structure of production, the allocation of capital, and the sustainability of long-term growth.

This study aims to analyze the impact of ESG factors on economic growth by integrating them into an extended Solow-type growth framework. While the expanded model traditionally emphasizes physical and human capital, we argue that ESG performance constitutes a distinct form of institutional and social capital that can improve total factor productivity and reduce systemic risks. The environmental, social, and governance indicators of the companies comprising the STOXX Europe 600 index over the period 2004–2023 are thus aggregated at the national level to construct a panel covering 16 European countries (Italy, United Kingdom, Germany, France, Belgium, Netherlands, Denmark, Finland, Spain, Portugal, Switzerland, Sweden, Poland, Austria, Ireland, and Norway).

The choice of the 2004–2023 period is particularly relevant in that it captures the transformation of ESG from a niche investment criterion to a central pillar of corporate strategy and European regulatory policy. The analysis of the countries represented in the STOXX Europe 600 also allows us to focus on advanced economies, where the conditional convergence predicted by the Solow model is more easily observable, particularly through institutional quality and sustainable investment. In this context, the relationship between ESG and growth is understood as a dynamic process that can be influenced by the level of economic development and the capacity to absorb structural transformations, in line with the Kuznets environmental curve hypothesis.

By combining the empirical rigor of the Mankiw–Romer–Weil model with contemporary sustainability indicators, this study contributes to the literature in several ways. First, it provides a macroeconomic assessment of the link between ESG performance and growth, whereas most existing work focuses on microeconomic analyses at the firm level. Second, the integration of ESG factors into an extended neoclassical growth framework enriches traditional models by incorporating the institutional, social, and environmental dimensions of sustainability. Finally, the use of panel methods that are robust to cross-sectional dependence, autocorrelation, and heteroscedasticity highlights the short-term effects of adjustment costs associated with the adoption of ESG criteria, interpreted in light of the Kuznets environmental curve hypothesis. These results contribute to a better understanding of the dynamic and potentially nonlinear nature of the relationship between sustainability and growth in advanced European economies.

2. Regulations implemented in Europe since 2000

Since the early 2000s, environmental, social, and governance (ESG) factors have undergone a fundamental transformation in the European context, evolving from a largely voluntary and normative framework into a series of institutionalized mechanisms integrated into economic and financial policies. Initially, sustainability initiatives were based on corporate social responsibility logic, relying on self-regulation and non-binding reporting frameworks. The European Sustainable Development Strategy adopted in the early 2000s exemplifies this initial phase, characterized by the political recognition of environmental and social issues without granting them legally binding status (European Commission, 2001).

However, this voluntary approach proved insufficient to ensure the homogeneous dissemination and effective comparability of non-financial information. In the absence of common standards and control mechanisms, ESG disclosure practices remained fragmented, limiting their usefulness for investors, researchers, and public authorities alike. This situation led to a growing recognition of the potential role of regulatory frameworks in structuring economic and financial sustainability at the European level.

The adoption of Directive 2014/95/EU on the disclosure of non-financial information (NFRD) marked a significant paradigm shift. This directive created a decisive turning point by imposing an obligation on large European companies to disclose information on environmental, social, and governance issues (European Parliament and Council of the European Union, 2014). By transforming the disclosure of ESG information from a voluntary practice into a legal requirement, the NFRD has contributed to increasing corporate transparency and laying the foundations for a more consistent data infrastructure across the European Union. As highlighted by Monciardini et al. (2020), this development has played a constructive role in restructuring the internal market information system by promoting greater corporate responsibility and better comparability of non-financial performance.

Beyond improving transparency, the NFRD has also changed the economic perception of ESG. By making this information more visible and systematically accessible, it has contributed to the gradual integration of sustainability issues into economic and financial decision-making processes. This development has paved the way for a more rigorous analysis of the links between ESG performance, capital allocation, and macroeconomic stability at both the microeconomic ( ) and aggregate levels.

As a continuation of these dynamics, the European Union began to embed sustainability more deeply at the heart of the financial system. The action plan for financing sustainable growth constituted a decisive step in this strategy by explicitly acknowledging that sustainability-related risks, particularly those related to climate change, must be fully assessed as financial risks (European Commission, 2018). This plan aims not only to redirect capital flows towards sustainable investments but also to strengthen the resilience of the European financial system against long-term environmental and social shocks.

The adoption of Regulation (EU) 2019/2088 on sustainability-related disclosures in the financial services sector (SFDR) is consistent with the logic of financializing ESG. By imposing greater transparency obligations on financial actors to integrate ESG risks into investment decisions, the SFDR has contributed to the internalization of sustainability considerations in capital allocation mechanisms (European Parliament and Council of the European Union, 2019). The classification of financial products according to sustainability commitments has also strengthened market discipline by limiting greenwashing practices and increasing the comparability of investment strategies.

In parallel, Regulation (EU) 2020/852, which defines the classification of sustainable activities, has introduced a common conceptual framework aimed at defining what constitutes sustainable economic activity (European Parliament and Council of the European Union, 2020). By providing a classification based on scientific criteria, the European taxonomy has reduced the uncertainty surrounding the concept of sustainability and facilitated the assessment of the environmental contributions of economic activities. This standardization has increased the reliability of ESG policies and ensured better integration of environmental considerations in investment and financing decisions.

The literature highlights that this regulatory development has contributed to reducing information asymmetries in financial markets and improving the assessment of intangible assets and long-term risks (Grewal & Serafeim, 2020). In this sense, the institutionalization of ESG contributes to a broader transformation of economic governance mechanisms by promoting a more efficient and potentially more stable capital allocation in the long term, beyond merely strengthening reporting obligations.

In the European context, this dynamic has been further strengthened by the adoption of the Corporate Sustainability Reporting Directive (CSRD) (EU) 2022/2464. The CSRD aims to improve the quality, comparability, and reliability of ESG data by expanding the scope of relevant companies and strengthening the standardization of published information (European Parliament and Council of the European Union, 2022). This development confirms that ESG is now fully integrated into European corporate frameworks and is one of the building blocks of long-term growth and stability strategies.

Despite these advances, the literature reveals that the macroeconomic impact of ESG is complex and potentially contradictory. The institutionalization of ESG criteria may encourage better allocation of capital, reduction of systemic risks, and increased efficiency in the long term, but it may also lead to transition costs in the short and medium term. These costs, linked to the adaptation of production structures and regulatory constraints, may exert temporary pressure on economic growth, particularly in economies facing structural rigidities.

In this context, analyzing ESG as a determinant of economic growth appears to be a natural extension of expanded growth models. The increasing availability of standardized ESG data and its incorporation into the regulatory framework now allows us to empirically examine, at the macroeconomic level, the extent to which sustainability policies affect the growth trends of European economies. Examining this relationship is particularly important in a context where sustainability and macroeconomic stability objectives are becoming increasingly intertwined in long-term economic policy strategies.

Literature Review

The literature on the determinants of economic growth has expanded steadily since the emergence of neoclassical growth models. Fundamental studies explained long-term economic performance primarily through physical capital accumulation, labor contributions, and demographic dynamics (Solow, 1957; Swan, 1956). Within this framework, technological progress was generally considered exogenous, and it was assumed that income level differences between countries would diminish over the long term. However, these models proved insufficient to explain the magnitude and persistence of empirically observed growth differences.

To overcome these limitations, subsequent studies enriched the analysis by integrating human capital, institutional quality, and endogenous technological progress as key determinants of economic growth (Mankiw, Romer, and Weil, 1992; Barro, 1991). These studies revealed the role of non-material factors in increasing total factor productivity and emphasized that economic growth depends not only on the accumulation of productive inputs but also on how these inputs are organized and mobilized within a specific institutional framework. From this perspective, it is now widely accepted that economic growth is shaped by the institutional and social context in which economic activities take place (North, 1990; Acemoglu, Johnson, and Robinson, 2005).

As an extension of this corporate approach, the literature has increasingly begun to focus on the interactions between environmental sustainability, governance structures, and economic performance. The hypothesis formulated by Porter and van der Linde (1995) has made an important contribution in this regard. Challenging the notion that environmental regulations necessarily hinder competitive strength, the authors argue that well-designed environmental policies can encourage innovation, increase resource efficiency, and ultimately improve economic performance. This approach suggests that compliance costs can be offset by increases in efficiency and technological advances, leading to a reassessment of the relationship between regulatory constraints and growth.

Empirically, much of the literature devoted to environmental, social, and governance criteria focuses on the microeconomic effects of ESG. Numerous studies show that companies with good ESG performance enjoy easier access to finance, lower capital costs, and greater resilience to economic shocks. A meta-analysis by Friede, Busch, and Bassen (2015), based on more than 2,000 empirical studies, found that a significant majority of studies concluded that there is a non-negative, even positive, relationship between ESG performance and companies' financial performance. These results suggest that integrating ESG criteria can be a means of creating value in the long term, rather than simply a regulatory constraint.

However, the literature also emphasizes that the economic impact of ESG depends on the nature and relevance of the issues addressed. Khan, Serafeim, and Yoon (2016) introduced the concept of financial materiality, showing that companies focusing on ESG issues that are material to their business areas outperform companies that adopt a general " " approach to sustainability. These results show that financial markets can distinguish between superficial practices, often referred to as "greenwashing," and concrete sustainability strategies, and tend to reward the latter.

At the same time, the shift of ESG from a voluntary framework to a regulated system has profoundly changed the nature of these practices. Grewal and Serafeim (2020) show that the institutionalization of ESG reporting, supported by changing investor preferences and increasingly restrictive regulatory frameworks, contributes to reducing information asymmetries in financial markets. The standardization of non-financial information strengthens the integration of sustainability into economic and financial decisions by enabling better assessment of intangible assets and long-term risks.

In the European context, this standardization process has gained momentum with the entry into force of the Non-Financial Reporting Directive (NFRD). Monciardini, Mähönen, and Tsagas (2020) emphasize that the transition from voluntary reporting to binding ESG disclosure requirements has profoundly restructured the information infrastructure of the European financial system. This development has not only increased the comparability and reliability of ESG data, but has also enabled a more systematic analysis of the interactions between sustainability, corporate governance, and macroeconomic stability.

Despite these advances, there is no clear consensus in the literature on the impact of ESG performance on economic growth at the macroeconomic level. While microeconomic results show positive effects, macroeconomic analyses reveal more heterogeneous outcomes. Many studies show that transition costs associated with environmental and social policies can put pressure on growth in the short term, especially in economies facing structural or institutional constraints. However, in the long term, the same policies can support more stable and sustainable growth through increased productivity, reduced systemic risks, and accelerated technological progress.

Overall, these factors indicate that the relationship between ESG performance and economic growth is complex, potentially non-linear, and largely dependent on the institutional context and level of economic development. This lack of consensus and the dominance of microeconomic analyses underscore the need for empirical approaches to systematically examine ESG's impact on macroeconomic growth, particularly in developed institutional frameworks such as European economies.

3. Research Methodology

Growth (GDP - Real GDP (constant 2017 national prices) and ESG data were obtained from the Bloomberg Refinitiv platform and OECD databases. In the study, GDP is used as the dependent variable, while ESG performance, human capital, physical capital stock, labor force, and average working hours are included as independent variables in the model. The panel data set consists of annual observations covering the period 2004–2023 for 16 European countries with companies listed on the STOXX Europe 600 index. Panel data analysis allows for the simultaneous examination of both the time series and cross-sectional dimensions (Yerdelen Tatoğlu, 2020).

The ESG data used in this study are derived from company-level ESG scores for companies included in the STOXX Europe 600 index. Company-level ESG scores are assigned to countries based on the country where the companies' headquarters are located. The country-level ESG indicator for each country and year was created by taking the simple arithmetic average of the ESG scores of companies operating in the relevant country. Missing observations regarding the ESG score were not included in the calculation for the relevant year. This approach aims to obtain comparable and consistent ESG indicators on a country basis.

4. The methodological position of aggregating company-level ESG indicators at the country level

The conversion of company-level environmental, social, and governance (ESG) indicators into country-level variables constitutes a central methodological step in this study. This process involves direct data creation choices, representativeness assumptions, and selected aggregation methods and should therefore be clearly presented in the section devoted to data and methodology, in line with the methodological standards of the empirical literature in economics (Wooldridge, 2010). Including this process in the introduction or literature review would be insufficient to meet the transparency and reproducibility requirements expected in a macroeconomic analysis based on aggregate data.

To this end, it is recommended that a specific subsection, such as "Aggregation of company-level ESG indicators at the country level," be added to the "Data and methodology" section. This subsection aims to explain the aggregation procedure used to derive national ESG indicators from microeconomic data, justify the selection of the company sample (specifically, companies comprising the STOXX Europe 600 index), and discuss the validity of these aggregated indicators as proxies for corporate and sustainability characteristics at the national level. This approach is widely adopted in the literature on sustainability and economic performance (Friede et al., 2015; Khan et al., 2016).

From the perspective of the reader and scientific reporter, this explanation is expected at the point when the variables are defined and included in the econometric model. In the absence of a specific methodological discussion, the use of ESG data obtained at the company level in macroeconomic analysis raises legitimate questions regarding collection biases, potential measurement errors, and the validity of the chosen proxy. This situation is highlighted in many recent studies in the literature on non-financial reporting and the standardization of ESG data (Monciardini et al., 2020; Grewal & Serafeim, 2020).

Furthermore, the position of this subsection within the methodology facilitates understanding of the analytical logic of the study. It establishes a clear link between microeconomic data sources, their conversion into macroeconomic indicators, and their use within the extended growth model, which is crucial for ensuring the internal consistency and empirical traceability of the analysis (Barro, 1991; Mankiw et al., 1992).

To ensure a smooth transition between the data definition and the econometric model presentation, the following introductory sentence can be added at the end of the subsection on data:

Since ESG indicators are initially measured at the company level, they must be aggregated at the country level to be used in a macroeconomic framework. The methodology chosen for this transformation is explained in detail in the next subsection.

This writing choice allows the ESG conversion to be naturally integrated into the methodological structure of the article and emphasizes its central role in the empirical analysis, consistent with the practices recommended in the applied econometric literature (Wooldridge, 2010).

5. Analysis

Data summaries of the studies conducted are provided below.

xtmixed lnrgdpna lnesgc lnrnna lnemp lnhc lnh

GDPNA: Real Gross Domestic Product

ESG: Environmental Social Governance

RNNA: Total Capital

EMP: Labor Force

HC: Human Capital

H: Annual Average Non-Working Hours (Leisure Time)

Table 1 Data summary

Variables	Observation	Average	St. Deviation	Min	Max
GDPNA	314	1394436	1323677	249,771.8	5112367
ESG	314	54.43906	10.97545	20.50956	78.00685
RNNA	314	7030972	6729014	968,077.3	2.13e+07
EMP	314	12.83419	12.47798	1.877763	46.03661
HC	314	3.28581	.3422848	2.230435	3.846449
H	314	2814.246	130.2533	2532.46	3066.43

 

The analysis includes 16 units across 20 time dimensions. First, logarithms were taken due to the very high difference between the max and min points of the variables.

The panel data regression was constructed as follows.

lnGDPNA_it= β₁ lnESG_it+ β₂lnRNNA_it+ β₃lnEMP_it+ β₄lnHC_it+ β₅lnH_it                                                                      (1)

i=1,….,N

t=1,…,T

In the panel data model shown in the regression model, N represents the unit dimension of the cross-section, while T represents the time dimension of the time series feature.

The panel data exhibits balanced panel characteristics. The panel analysis was tested for unit and time effects, and it was found to have both unit and time effects. The time effect was controlled for by assigning shadow variables, and the random effects model was preferred in line with the Hausman test.

The fixed effects and random effects models were tested separately. They were tested using the Hausman test.

 

lnGDPNA_it  =β₁ lnESG_it + β₂ lnRNNA_it + β₃ lnEMP_it + β₄ lnHC_it + β₅ lnH_it  +μ _it  + e _it                                           (2)

lnGDPNA_it  =β₁ lnESG_it + β₂ lnRNNA_it + β₃ lnEMP_it + β₄ lnHC_it + β₅ lnH_it + e _it                                                      (3)

 

Hausman Test Statistic;

H = (^βSE-^βTE)’[Var(^βSE)-Var(^BTE)]^-1 (^βSE-^βTE)

The H statistic has degrees of freedom equal to the number of coefficients in the fixed and random effects model.

It follows a chi^-squared distribution. In the Stata output;

Table 2 Chi-squared distribution

Test: Ho: difference in coefficients not systematic
chi2(5)	(b1-b2)' * [V_bootstrapped(b1-b2)]^(-1) * (b1-b2)
=	0.24
Prob>chi2	0.9986

 

Since the probability value is greater than 0.05, it indicates that the model is consistent with the random effects model.

The model that fits the random effects model was tested for econometric assumptions.

First, the presence of multicollinearity among the independent variables in the model was tested. Multicollinearity refers to the existence of a high level of linear relationship among the explanatory variables. If this problem exists, the parameter estimates obtained using the Least Squares (LS) method may lose their reliability; negative consequences may arise, such as coefficients not reflecting reality, expected signs reversing, variances increasing, and consequently wide confidence intervals (Örk Özel & Gezer, 2020).

The Variance Inflation Factor (VIF) criterion was used to detect multicollinearity. Linearity refers to two independent variables that are nearly linear combinations of each other. Multicollinearity occurs when there are several variables in the regression model that are significantly related not only to the dependent variable but also to each other. (Young, 2017) This situation can lead to misleading or erroneous results when the researcher attempts to evaluate the explanatory or predictive power of each variable separately. In general, multicollinearity increases the standard errors of coefficient estimates, leading to wider confidence intervals and weaker statistical significance levels. Therefore, findings from models with multicollinearity may not be reliable. (Frank, 2001)

Here, the tolerance value is equal to the inverse of the Variance Inflation Factor (VIF). As the tolerance value decreases, the likelihood of multicollinearity between variables increases. A VIF value of 1 indicates that there is no linear relationship between the independent variables. Values in the range 1 < VIF < 5 indicate a moderate level of correlation between the variables. A VIF value between 5 and 10 is considered critical as it indicates a high degree of correlation between the variables. When VIF ≥ 5–10, multicollinearity problems arise in the regression model; VIF > 10 indicates that regression coefficients are estimated weakly and unreliably due to multicollinearity. (Belsley, 1991)

Table 3 Multicollinearity

Variables	VIF	1/VIF
LRNNA	10.44	0.095795
LEMP	10.27	0.097324
LH	1.46	0.685083
LHC	1.40	0.714169
ElSG	1.14	0.875077
Average VIF	4.94

 

Based on the results obtained, the average VIF value was calculated as 4.94. The fact that this value is below 5 indicates that there is no serious multicollinearity problem in the model as a whole. However, although the VIF values for some variables were observed to be relatively high, this situation is not considered to compromise the reliability of the coefficient estimates.

Another test for deviation from assumptions is the test for heteroscedasticity. It states that the conditional variance of the error term, which cannot be observed under the independent variable condition, is constant for all observations. In other words, the variance of the error term does not show a systematic change depending on the level of the independent variables. As emphasized by Wooldridge, heteroscedasticity arising from a violation of this assumption does not bias the Ordinary Least Squares (OLS) coefficient estimates but leads to inconsistent standard errors, thereby weakening the validity of statistical inferences based on traditional t and F tests. ( In the random effects model, the Levene, Brown, and Forsythe Test was used to test for heteroscedasticity. The null hypothesis of the test is homoscedasticity, while the alternative hypothesis is the presence of heteroscedasticity. The test concluded that there was no heteroscedasticity. When testing for deviations from the assumption, autocorrelation was tested using the Durbin-Watson and Baltagi-Wu tests. Test statistics:

 

Table 4 Autocorrelation test

Modified Bhargava et al. Durbin–Watson	0.32932423
Baltagi–Wu LBI	0.48912354

 

Accordingly, values less than 2 indicate the presence of autocorrelation. Inter-unit correlation in the model was tested using the Pesaran Test and the Friedman Test. Tests based on the null hypothesis of no inter-unit correlation concluded that the model exhibits inter-unit correlation.

 

Table 5 Inter-unit correlation

Pesaran's test of cross-sectional independence	9.683	Pr = 0.0000
Friedman's test of cross-sectional independence	61.239	Pr = 0.0000

 

The model's suitability for normal distribution was tested using the D'Agostino, Belanger, and D'Agostino Test. It was concluded that the model's error terms were normally distributed at a 95% significance level, while the unit effect was normally distributed at a 90% significance level.

 

6. Model estimation

In cases of heteroscedasticity, autocorrelation, and inter-unit correlation, the variance-covariance matrix of the error terms is not equal to the product of the residual variance and the unit matrix; in other words,

E(u(t)) ≠ Q²uI_t, therefore the equality E(u_tu’_t) = Q’_u Ω_T holds.

In this case, when there is no heteroscedasticity, autocorrelation, or cross-correlation,

Var(β^)=E[(X’X)^-1 X’uu’X(X’X)^-1  ]

This situation does not cause inconsistency when working with large samples, but it does affect efficiency. In other words, the validity of the parameter variances and, consequently, the standard errors, the t and F statistics, R², and the confidence intervals are affected. Therefore, if any of heteroscedasticity, autocorrelation, or inter-unit correlation is present in the model, either the standard errors should be corrected without touching the parameter estimates, which is possible with robust estimation, or estimates should be made using appropriate methods if they are present. (Yerdelen Tatoğlu, 2020;303)

 

Table 5 Error processes and corresponding robust estimators

Huber (1967), Eicker (1967), and White(1980) Estimator	Heteroskedasticity
Arellano (1987), Froot (1989), andRogers (1993) Estimator	Heteroscedasticity and Autocorrelation
Driscoll Kraay (1998) Estimator	Heteroscedasticity, Autocorrelation, and Inter-unit Correlation
AR(1) Residual Linear RegressionModel	First-Order Autocorrelation

 

The model was estimated using generalized least squares (GLS), and standard errors were corrected for heteroscedasticity, autocorrelation, and cross-dependency using the Driscoll–Kraay (1998) method.

Table 6 GLS regression results with Driscoll–Kraay standard errors

Explanatory Variables	Coefficients	Driscoll KraaySt. Error	T-Statistic Probability
lnGDPNA	9.607751	1.864274	0
lnESG	-.0669943	.0103371	0.0
lnRNNA	.9718476	.0444616	0.0
lnEMP	.021674	.0486669	0.661
lnHC	-.0051569	.0745333	0.946
lnH	-1.326676	.2840688	0
Diagnostic Tests		Result
Wald x2Test		0.0
R2		0.92
Maximum delay		2

 

In the model where the maximum lag number was set to 2, Driscoll Kraay was used to eliminate autocorrelation and inter-unit correlation. Looking at R², which takes values between 0 and 1 and is the ratio of independent variables explaining the dependent variable, it is seen that it is 0.92, and it can be said that this ratio is high.

Since the Wald x²Test Statistic probability value is 0.0, the model is considered significant.A 1% increase in the ESG score negatively affects gross domestic product by 0.06%, while a 1% increase in leisure time negatively affects gross domestic product by 1.32%. A 1% increase in capital stock positively affects gross domestic product by 0.97%, and a 1% increase in labor positively affects gross domestic product by 0.02% ( ). In addition, the model estimation found that a 1% increase in human capital (HC) changes gross domestic product by a very small amount.

7. Conclusion

This study empirically analyzes the relationship between environmental, social, and governance (ESG) performance and economic growth within a macroeconomic framework, based on a panel of 16 European countries for the period 2004-2023. The results, which integrate ESG indicators into an extended neoclassical growth model of the Solow-Mankiw-Romer-Weil type, reveal a statistically significant relationship between sustainability factors and growth dynamics. Estimates show that physical capital accumulation remains the main driver of growth, while the ESG indicator has a negative short-term effect on real gross domestic product. The results are robust to heteroscedasticity, autocorrelation, and cross-dependency issues thanks to the use of Driscoll–Kraay standard errors.

However, the negative impact of ESG performance on growth should not be interpreted as questioning the validity of sustainability policies. This effect reflects the existence of adjustment costs associated with the transition to more sustainable production models. In advanced European economies, strengthening environmental, social, and governance standards requires significant investments in regulatory compliance, technological adaptation, and corporate restructuring, which may temporarily slow economic growth. This dynamic is consistent with Kuznets' environmental curve hypothesis. According to this hypothesis, the initial and intermediate stages of tightening sustainability standards may have temporary negative effects on growth before the gains related to innovation, production efficiency, and reduced systemic risks materialize in the long term.

From this perspective, the negative coefficient associated with ESG reveals a time arbitrage between short-term transition costs and the long-term potential benefits of sustainability policies. The results show that the macroeconomic impact of ESG is largely dependent on the level of economic and institutional development and the capacity of economies to absorb and internalize the structural transformations brought about by the sustainable transition. Therefore, ESG criteria appear to be a factor whose effects vary depending on the time frame and institutional context, rather than a structural barrier to growth.

The ESG variable used in this study is not considered as a sustainability indicator for the entire national economy, but rather as a proxy for corporate, environmental, and social practices observed in large-scale, publicly traded companies. In contrast, the human capital variable is a macro-level indicator representing the overall education and skill level of the economy. These different measurement levels are consistent with the assumption that ESG practices emerging through large firms may affect the production process across the country through technology diffusion and productivity externalities, and do not undermine the theoretical consistency of the model.,

The ESG variable used in this study is not considered as a sustainability indicator for the entire national economy, but rather as a proxy for corporate, environmental, and social practices observed in large-scale, publicly traded companies. In contrast, the human capital variable is a macro-level indicator representing the overall education and skill level of the economy. These different measurement levels are consistent with the assumption that ESG practices emerging through large firms may affect the production process across the country through technology diffusion and productivity externalities, and do not undermine the theoretical consistency of the model. Finally, these results contain important implications for public policymakers. They emphasize that ESG policies should be supported by measures aimed at reducing transition costs, particularly by supporting innovation, productive investment, and human capital development. For future research, investigating non-linear relationships, development thresholds, or differing effects among environmental, social, and governance components could provide a better understanding of the complex link between sustainability and economic growth in advanced economies.

References

1. Acemoglu, D., Johnson, S., & Robinson, J. A. (2005). Institutions as a fundamental cause of long-run growth. In P. Aghion & S. N. Durlauf (Eds.), Handbook of Economic Growth (Vol. 1A, pp. 385–472). Elsevier. https://doi.org/10.1016/S1574-0684(05)01006-3
2. Alogoskoufis, S., Dunz, N., Emambakhsh, T., Hennig, T., Kaijser, M., Kouratzoglou, C., Muñoz, M. A., Parisi, L., & Salleo, C. (2021). ECB economy-wide climate stress test: Methodology and results (Occasional Paper Series No. 281). European Central Bank.
   https://doi.org/10.2139/ssrn.3929178
3. Barro, R. J. (1990). Government spending in a simple model of endogenous growth. Journal of Political Economy, 98(5, Part 2), S103–S125. https://doi.org/10.1086/261726
4. Belsley, D. A. (1991). Conditioning diagnostics: Collinearity and weak data in regression. Wiley. ISBN 9780471528890.
5. Busch, T., Bruce-Clark, P., Derwall, J., Eccles, R., Hebb, T., Hoepner, A., Klein, C., Krueger, P., Paetzold, F., Scholtens, B., & Weber, O. (2021). Impact investments: A call for (re)orientation. SN Business & Economics, 1, Article 33. https://doi.org/10.1007/s43546-020-00033-6
6. Eccles, R. G., Ioannou, I., & Serafeim, G. (2014). The impact of corporate sustainability on organizational processes and performance. Management Science, 60(11), 2835–2857. https://doi.org/10.1287/mnsc.2014.1984
7. Friede, G., Busch, T., & Bassen, A. (2015). ESG and financial performance: Aggregated evidence from more than 2,000 empirical studies. Journal of Sustainable Finance & Investment, 5(4), 210–233. https://doi.org/10.1080/20430795.2015.1118917
8. Grewal, J., & Serafeim, G. (2020). Research on corporate sustainability: Review and directions for future research. Foundations and Trends in Accounting, 14(2), 73–127. https://doi.org/10.1561/1400000061
9. Harrell, F. E., Jr. (2001). Regression modeling strategies: With applications to linear models, logistic regression, and survival analysis. https://doi.org/10.1007/978-1-4757-3462-1_10
10. Khan, M., Serafeim, G., & Yoon, A. (2016). Corporate sustainability: First evidence on materiality. The Accounting Review, 91(6), 1697–1724. https://doi.org/10.2308/accr-51383
11. Mankiw, N. G., Romer, D., & Weil, D. N. (1992). A contribution to the empirics of economic growth. The Quarterly Journal of Economics, 107(2), 407–437. https://doi.org/10.2307/2118477
12. Monciardini, D., Mähönen, J., & Georgin, M. (2020). Rethinking non-financial reporting: A blueprint for structural regulatory changes. Accounting, Economics, and Law: A Convivium, 10(2). https://doi.org/10.1515/ael-2020-0092
13. North, D. C. (1990). Institutions, institutional change and economic performance. Cambridge University Press. https://doi.org/10.1017/CBO9780511808678
14. Örk Özel, S., & Gezer, F. (2020). Comparison of estimators under multicollinearity problem: Generalized maximum entropy, ridge, liu. Dumlupınar University Journal of Social Sciences, (66), 82–96.
15. Porter, M. E., & van der Linde, C. (1995). Toward a new conception of the environment-competitiveness relationship. Journal of Economic Perspectives, 9(4), 97–118. https://doi.org/10.1257/jep.9.4.97
16. Solow, R. M. (1957). Technical change and the aggregate production function. The Review of Economics and Statistics, 39(3), 312–320. https://doi.org/10.2307/1926047
17. Swan, T. W. (1956). Economic growth and capital accumulation. The Economic Record, 32(2), 334–361. https://doi.org/10.1111/j.1475-4932.1956.tb00434.x
18. Wooldridge, J. M. (2004). Introductory econometrics: A modern approach (2nd ed.). South-Western (Cengage).
19. Yerdelen Tatoğlu, F. (2020). Panel veri ekonometrisi: Stata uygulamalı (5th ed.). Beta Yayınları (Istanbul).
20. Young, D. S. (2017). Handbook of regression methods. CRC Press. ISBN 9781498775298.