The Role of Economic Freedom and Institutional Quality in Driving Sustainable Development: Comparative Evidence from Developed and Developing Economies

Beginning in the 19th century, the world has witnessed numerous disruptions across economic, political, and social spheres. Major events such as the 1997 Asian financial crisis, the bursting of the dotcom bubble in 2000, the global financial meltdown in 2008, the COVID-19 outbreak in 2019, and recent geopolitical tensions like the Russia–Ukraine conflict in 2022 and the Hamas–Israel war have all exerted considerable pressure on the global economy. Beyond their financial consequences, these crises have underscored systemic global vulnerabilities, including escalating environmental pollution, resource exhaustion, climate change, and issues related to energy and food security. These challenges are closely tied to long-standing patterns of industrial activity and scientific advancement driven by human development over centuries. A prevailing focus on GDP expansion—often pursued without sufficient consideration for ecological impacts—has accelerated the depletion of natural assets, degraded ecosystems, and undermined environmental stability [1]. As a result, growing environmental deterioration has sparked global alarm over the feasibility of achieving sustainable development. Today, there is widespread consensus that genuine progress requires balancing the three foundational pillars of sustainability: economic viability, social well-being, and environmental preservation [2-5].

However, achieving SDGs is challenging due to environmental pollution and climate change, primarily caused by the industrialization process that is inseparable from continuous production [6]. The situation is even more complicated in developing countries, where economic and social development, poverty alleviation, and improving living standards are closely linked to economic freedom (EFI) [7]. Krueger [8] argued that a market is considered free if there are no quantity constraints to control both buyers and sellers. Thus, deregulation can be seen as the action of removing quantity controls. Broadly speaking, deregulation is considered any policy aimed at reducing state control, or completely eliminating it, or replacing it with a more relaxed form of control. There are numerous studies on the impact of economic freedom on economic growth today [9-11]. However, these studies often overlook environmental and social issues – issues related to sustainable development. Furthermore, in developing countries with low per capita income, they often focus more on economic development goals rather than environmental quality. In other words, the institutions of these countries often encourage growth and sacrifice the environment by focusing on "brown industries" that cause pollution. Conversely, in developed countries, although there is infrastructure for applying technologies to minimize environmental pollution, some countries are still on the list of environmentally polluting countries, including the United States [12], which remains an open question. One of the reasons is population growth. Therefore, the impact of economic freedom on sustainable development cannot ignore the role of the institutional quality (IQ) of countries.

Current studies on the impact of economic freedom on sustainable development are still very limited [13, 14]. Recently, Lima et al. [15] studied economic privacy and sustainable development, considering the role of circular economy, while Shahnazi and Shabani [16] studied the impact of economic freedom on sustainable development, considering the role of renewable energy. Despite a growing body of literature on economic freedom and development, several important research gaps remain.

First, most prior studies analyze only one or two dimensions of sustainable development—economic growth, environmental quality, or social progress—without capturing the full scope of sustainability. This study addresses this gap by employing the SDGI, which incorporates all 17 SDGs.

Second, although some research spans global or regional scales, few studies distinguish between developed and developing countries, despite their divergent institutional frameworks and sustainability priorities. This study fills that contextual gap by offering a comparative analysis between the two country groups.

Third, previous empirical works predominantly rely on frequentist methods that impose strict assumptions and treat model parameters as fixed. In contrast, this research applies Bayesian regression to offer more robust probabilistic inferences, particularly valuable in settings with limited or noisy data.

By addressing these three gaps, the study contributes novel insights into how economic freedom and IQ jointly shape sustainable development in varying contexts.

The structure of the research consists of five parts. The second part discusses the previous related literature, while the third part details the data and methodology. The findings and experimental discussions are presented in part four, and the conclusions and policy implications are presented in part five.

3.1 Research models and data

To evaluate the relationship between economic freedom and sustainable development, this study adopts a quantitative research approach. Following the methodology proposed by Oanh and Dinh [3] and Van and Le Quoc [4], the Sustainable Development Goal Index (SDGI) is computed by aggregating 17 indicators (see Appendix 1), each corresponding to one of the three core pillars of sustainability: economic, social, and environmental, as defined in the SDGs. This composite index is widely regarded as a comprehensive measure of a nation’s sustainable development performance [37], with data sourced from the official sdgindex.org portal.

The EFI is formulated by integrating several economic and institutional variables that reflect ethical business conduct and the regulatory environment. Key components include property rights, judicial independence, government transparency, fiscal burden, public expenditure, macroeconomic stability, and freedoms related to business, labor, money, trade, investment, and finance. The EFI score is derived as the arithmetic mean of these indicators, where values range from 0 (no economic freedom) to 100 (maximum freedom). This metric has been extensively used in earlier empirical works [14, 38] with data obtained from the Heritage Foundation.

To capture IQ, the study utilizes the Worldwide Governance Indicators (WGI) compiled by the World Bank’s World Development Indicators (WDI) database. These indicators assess a country's governance capacity across six dimensions: Voice and Accountability (VOA), Political Stability and Absence of Violence (POL), Government Effectiveness (GOV), Regulatory Quality (REG), Rule of Law (RUL), and Control of Corruption (COR). Each dimension is scored between -2.5 and 2.5. Although incorporating all six variables as regressors could provide a comprehensive portrayal of institutional performance [39], this approach may introduce multicollinearity due to potential overlap among the indicators. Additionally, including all components could lead to model overfitting and reduce statistical efficiency. To address these concerns, this study adopts Principal Component Analysis (PCA) to consolidate the governance indicators into a single index that robustly reflects IQ, following the approach of Ullah and Khan [40].

As emphasized by North [18] and Kasper et al. [21], institutions shape the rules of the game in economic exchange, and their quality determines how effectively economic freedom can function without leading to adverse consequences. In countries with weak IQ—where enforcement of environmental laws is limited and corruption is prevalent—economic freedom may inadvertently amplify pollution and unsustainable practices, as firms exploit lax regulation to maximize short-term profits. Conversely, in institutional environments with high accountability and strong RUL, economic freedom can foster innovation, encourage clean technologies, and enhance sustainable investment. Therefore, this study incorporates an interaction term between EFI and IQ to capture this moderating effect, reflecting the idea that the impact of economic freedom on sustainability is contingent upon institutional context.

Moreover, the selected control variables included in the model encompass economic growth, environmental quality, foreign direct investment (FDI), trade openness (OPEN), urbanization rate (UR), population growth rate (POP), and inflation rate (INF). The source and data of the research are presented in Appendix 1. The research model is as follows:

$\begin{gathered}S D G I_{i, t}=\beta_o+\beta_1 E F I_{i, t}+\beta_2 I Q_{i, t}+\beta_2 E F I * I Q_{i, t}+\beta_x Z_{i, t}+\varepsilon_{i, t} \\ \text { For } \mathrm{i}=1,2, \ldots, \mathrm{n} ; \mathrm{t}=1,2, \ldots, \mathrm{t}\end{gathered}$     (1)

where, i represents countries and t represents observation time points in the model from 2008 to 2022. The variable Z represents the control variables including: (1) CO₂ emissions (CO₂); (2) UR; (3) FDI; (4) POP; (5) GDP growth (GDP), (6) INF, (7) Financial development index (FD), and (8) OPEN.

UR: The share of the population living in urban areas reflects the level of urban development. Urbanization can support sustainability through better infrastructure, healthcare access, education, and technology adoption. As noted by Van and Quoc [4], urbanization positively influences sustainable development, especially in countries where public services are concentrated in urban centers.

FDI: FDI plays a dual role in sustainability. It can promote technological transfer, infrastructure development, and job creation, contributing positively to SD. However, in countries with lax regulations, it may also lead to "pollution havens" by relocating environmentally harmful industries. Due to this duality, we do not impose a directional assumption on its effect.

POP: Rapid population growth may exert pressure on public infrastructure, natural resources, and environmental quality. It can also lead to urban overcrowding and increased energy consumption, thereby constraining sustainable development. Hence, we expect a negative relationship between population growth and sustainability.

GDP Growth (GDP): Economic growth expands the fiscal space and investment capacity of governments, enabling higher spending on sustainability initiatives. However, unchecked growth—especially when driven by pollution-intensive sectors—may conflict with environmental goals. Therefore, we do not specify the expected sign of GDP growth in this study.

INF: High inflation is often associated with macroeconomic instability and inefficient resource allocation, especially in sectors essential for long-term sustainability such as healthcare and green technology. As highlighted by Van and Quoc [4], inflation tends to negatively affect sustainable development.

Financial Development Index (FD): Financial development improves the efficiency of capital allocation, enhances access to credit, and facilitates green investment. A well-developed financial system also supports innovation and sustainability-related entrepreneurship. Therefore, we expect a positive relationship between financial development and sustainable development [4].

OPEN: Trade openness allows countries to access advanced technologies and environmentally friendly practices, potentially supporting sustainable growth. However, it can also result in environmental degradation if a country becomes a destination for dirty industries. Due to this ambivalence, we do not predetermine the direction of the effect [4].

CO₂ Emissions (CO₂): CO₂ emissions per capita serve as a proxy for environmental degradation. Higher emissions are generally associated with increased industrial activity and fossil fuel consumption, which negatively affect environmental sustainability. Numerous studies (e.g., Adesina and Mwamba [31]; Wu et al. [34]) have demonstrated a significant negative relationship between CO₂ emissions and sustainable development outcomes.

3.2 Methodology

Bayesian statistics integrates empirical data with prior beliefs to generate a posterior distribution, which is interpreted as the probability distribution of parameter values. Unlike traditional methods, Bayesian inference does not heavily depend on large sample sizes, making it particularly suitable for small-sample studies [41-45]. Fundamentally, Bayesian and frequentist approaches are grounded in distinct philosophical frameworks. In Bayesian thinking, the observed data is treated as fixed, while the model parameters are viewed as random variables. In contrast, frequentist methods consider the data as arising from repeated sampling, with the parameters regarded as unknown but constant. Bayesian inference estimates the posterior distribution of parameters by combining the likelihood function derived from observed data with a prior distribution. This is expressed formally as:

$y \sim N\left(\beta^T X, \sigma^2 I\right)$     (2)

where, y is generated from a normal distribution characterized by the mean and variance. The mean value of the linear regression is the transpose of the weight matrix multiplied by the predictor matrix. The variance is the square of the standard deviation (σ) multiplied by the identity matrix.

Not only the output (y) is generated from a probability distribution, but also the model parameters are assumed to come from a distribution. The posterior probability of the model parameters conditioned on the inputs and outputs takes the following form:

$P(\beta \mid y, X)=\frac{P(y \mid \beta, X)(P(\beta \mid X))}{P(y \mid X)}$     (3)

where, P(β∣y,X) is the posterior probability distribution of the model parameters for the inputs and outputs; P(y,X) is the likelihood of the data; P(β∣X) is the prior probability distribution, and P(y∣X) is the normalization constant and can be dropped. Therefore, equation (**) is often simplified to:

$P(\beta \mid y, X)=P(y \mid \beta, X)(P(\beta \mid X))$     (4)

Regarding prior specification, this study employs weakly informative normal priors centered at zero for the regression coefficients. This reflects a conservative assumption that the parameters are more likely to be close to zero in the absence of strong empirical signals. Such priors help minimize prior-driven bias and are widely used in Bayesian regression modeling when the goal is to allow the data to dominate inference [46-52]. For variance parameters, standard inverse-gamma priors are used. To estimate the posterior distributions, the model applies Markov Chain Monte Carlo (MCMC) simulation using the Metropolis–Hastings algorithm. To ensure the reliability of the simulations, we monitor the Gelman–Rubin convergence diagnostic (Rc). This statistic compares the variance between multiple MCMC chains to the variance within each chain. A Rc value close to 1.0000 indicates strong convergence, confirming that the chains have mixed well and that posterior estimates are stable. This approach ensures that the Bayesian model is both theoretically sound and computationally reliable.

This study investigates how economic freedom influences sustainable development, incorporating IQ as a moderating factor. The analysis is based on data from 2008 to 2023, covering 22 developing and 26 developed countries. Traditional frequentist approaches in macroeconomic research often face challenges such as multicollinearity, autocorrelation, and issues related to small sample sizes. To address these limitations, this study employs a Bayesian regression framework, which offers greater robustness against common econometric issues including heteroscedasticity, endogeneity, and limited data.

A notable advantage of the Bayesian approach is its ability to provide probabilistic interpretations of coefficient estimates. The findings reveal contrasting dynamics between developing and developed countries. In developing nations, both economic freedom (EFI) and IQ negatively affect sustainable development, with posterior probabilities of 100% and 98.6%, respectively. These results suggest that in these countries, economic policies and institutional frameworks may prioritize industrial growth at the expense of environmental sustainability. However, the interaction between EFI and IQ yields a positive influence, indicating that coordinated efforts between economic freedom and institutional policies can lead to more sustainable outcomes.

Conversely, in developed economies, EFI and IQ show a positive impact on sustainable development, while their interaction is associated with a negative effect. This supports the perspective that institutional mechanisms and market freedoms may function more effectively when operating independently in these contexts.

The study also highlights that in both country groups, variables such as GDP, CO₂ emissions, and population (POP) are negatively associated with the SDGI, whereas unemployment (UR) and inflation (INF) exhibit a positive correlation with SDGI.

Bayesian model diagnostics further confirm the reliability of the results. The lowest average acceptance rates for the Metropolis-Hastings algorithm are 0.9578 (developing countries) and 0.9788 (developed countries), both exceeding the commonly accepted threshold of 0.1. Furthermore, the minimum efficiency values recorded are 0.7554 and 0.8863, respectively—well above the benchmark of 0.01. All parameter chains exhibit Rc values equal to 1, confirming convergence of the MCMC process. These indicators collectively affirm the robustness and reliability of the Bayesian simulation results.

For Developing Countries

Strengthen IQ to support environmental regulation: Governments should enforce stricter rules on energy use, waste disposal, and pollution control to mitigate environmental harm. Clear and enforceable regulations can create a healthier living environment and align economic growth with sustainability.

Promote green innovation through institutional incentives: Institutions should facilitate access to finance, technology, and infrastructure for green industries. Encouraging a transition to renewable energy—despite its upfront costs—requires targeted policies and institutional support for technological innovation.

Ensure fair market competition and accountability: Economic freedom must be grounded in transparent, rule-based systems. Governments should enact and enforce clear competition laws, internalize environmental costs (e.g., pollution taxes), and discourage monopolistic behavior to foster dynamic and sustainable markets.

For Developed Countries

Maintain the independence between institutions and market freedom: In mature economies, excessive institutional interference in markets may hinder efficiency. Policymakers should allow economic freedom to operate independently while maintaining strong legal frameworks for oversight.

Prioritize public-sector investment in sustainability innovation: Institutions should focus on funding research, infrastructure, and public-private partnerships that accelerate clean technologies, sustainable production, and circular economy models.

Promote balanced growth through institutional coordination: While minimizing overregulation, institutions should still guide the private sector toward sustainable practices through incentives, environmental standards, and long-term strategic planning.

For Developed Countries

Institutions should be independent of economic freedom. In a strong free-market economy, the market often determines business and consumer behavior. In some cases, this can lead to conflicts of interest, such as businesses focusing on maximizing profits without considering the environmental or social impact. Therefore, the independence of economic freedom and institutions in developed countries is necessary. Furthermore, the institutions of developed countries should focus on public sector, where they can provide financial support and collaborate with research organizations to promote innovation and industrial development, enhance sustainable environments, and foster the development of the private sector to promote sustainable economies.

One important limitation of this study lies in its use of the aggregate SDGI as a proxy for a country’s overall level of sustainable development. Although the SDGI is a comprehensive indicator encompassing progress across all 17 SDGs, relying on a composite index may obscure the nuanced differences among the three individual pillars of sustainable development—economic, social, and environmental. For instance, a country might perform well in achieving economic-related SDGs while lagging significantly in environmental or social goals. The use of a single aggregate measure could mask these imbalances, potentially leading to less precise conclusions about the specific influence of economic freedom and IQ on each aspect of sustainability. Therefore, future research is encouraged to disaggregate the SDGI into its constituent components—such as separate indices for economic, social, and environmental sustainability—or to examine selected SDGs individually. This would allow for a more detailed and targeted assessment of how economic freedom and governance institutions affect different dimensions of sustainable development.