The paper addresses the growing concerns about the interplay between economic growth and environmental sustainability, particularly in light of rising trade protectionism. The Environmental Kuznets Curve (EKC) hypothesis, suggesting an inverted U-shaped relationship between economic growth and environmental degradation, is central to the discussion. However, the recent increase in trade protectionism, fueled by events like the 2008 financial crisis, the COVID-19 pandemic, and the Russia-Ukraine war, introduces significant complexities. The study focuses on understanding how trade protection, measured by trade openness, influences the EKC. This is crucial because the impact of trade on the environment is debated, with the 'pollution paradise' hypothesis suggesting that trade leads to environmental degradation in developing countries (by shifting polluting industries) and the 'pollution halo' hypothesis suggesting that it can improve environmental quality in recipient countries (through technology transfer). The research aims to answer three key questions using panel data from 147 countries (1995-2018): 1) Does the EKC hypothesis hold within the study's scope? 2) What is the role of trade protection in the economic growth-environmental degradation relationship, and are there non-linear effects? 3) How do the observed effects differ across income groups?

The literature review extensively examines existing research on the EKC hypothesis, exploring diverse methodologies and models used to analyze the relationship between economic growth and environmental degradation (using indicators like CO2 emissions and ecological footprint). The review highlights studies supporting the inverted U-shaped EKC relationship, alongside studies questioning the traditional EKC and proposing alternative relationships (N-shaped, inverted N-shaped). The review also explores the existing literature on the impact of trade on the environment, contrasting the 'pollution paradise' and 'pollution halo' hypotheses. It notes the lack of direct research on trade protection's impact on the environment, with most studies using trade openness as a proxy for trade liberalization. The review concludes by emphasizing the need for a systematic examination of the trade-economy-environment nexus and a re-evaluation of the EKC from a trade perspective, incorporating both carbon emissions and ecological footprint as environmental indicators and comparing results across different income levels.

The study uses panel data for 147 countries from 1995 to 2018, with data sourced from the World Bank, Global Footprint Network, and KOF Swiss Economic Institute. The data includes per capita GDP (as the explanatory variable), trade openness (as the threshold variable), per capita carbon emissions and ecological footprint (as explained variables), and control variables like industrial structure, foreign direct investment, and the KOF Globalization Index. The data is logarithmically transformed to mitigate outliers and multicollinearity issues. A panel unit root test (LLC and IPS) ensures data stationarity. A Pedroni cointegration test checks for long-term relationships between variables. The core of the analysis employs a nonlinear threshold panel regression model (Hansen, 1999), using trade openness as the threshold variable. This model captures the non-linear effects of trade on the relationship between economic growth and environmental degradation. Both single and double threshold models are estimated and tested for significance using the Bootstrap method. The study also conducts separate analyses for four income groups (based on World Bank classifications) to explore heterogeneity in the effects of trade protection across different development levels.

The study's key findings support the EKC hypothesis, showing an inverted U-shaped relationship between economic growth and environmental degradation (carbon emissions and ecological footprint). However, the relationship is influenced by trade protection. For the global panel, the double threshold effect of trade on the economic growth-carbon emission model is significant at the 10% level, with thresholds at 4.188 and 5.019. Similarly, the double threshold effect is significant at the 10% level for the economic growth-ecological footprint model, with thresholds at 2.979 and 5.540. Analysis of the four income groups reveals significant double-threshold effects for carbon emissions in all groups (significant at 5% for high and lower middle-income groups, 10% for others). For ecological footprint, the high-income group shows a significant single threshold effect, while the low and lower-middle-income groups show significant double threshold effects. The threshold values vary significantly across income groups. Control variables like industrial structure and globalization index show significant positive effects on both carbon emissions and the ecological footprint. High-income countries show the lowest positive effect of economic growth on environmental degradation, while lower-middle-income countries show the highest. This demonstrates an EKC pattern across the income groups. Trade protection exhibits a nuanced impact, mitigating environmental degradation in high-income countries but exacerbating it in other income groups, lending support to the pollution haven hypothesis.

The findings address the research questions by confirming the EKC hypothesis within the study's scope, demonstrating the non-linear influence of trade protection on the environment-economy relationship, and highlighting heterogeneity across income groups. The EKC's validity underlines the potential for economic growth to eventually lead to environmental improvements, contingent on reaching a certain development level and implementing effective policies. The non-linear effect of trade protection emphasizes the need for nuanced policy responses that consider both economic growth and environmental sustainability. The support for the pollution haven hypothesis suggests that trade liberalization may disproportionately benefit high-income countries while potentially exacerbating environmental problems in developing economies. The varying impact across income groups further demonstrates the importance of considering regional heterogeneity when designing and implementing environmental policies.

The study makes several key contributions: it validates the EKC from a new perspective incorporating trade protection, utilizes a rigorous nonlinear threshold panel regression model to analyze complex relationships, and reveals significant heterogeneity in the effects of trade across different income levels. Future research could expand the dataset geographically and temporally, explore additional variables (e.g., renewable energy consumption, environmental regulations), and employ more sophisticated econometric techniques.

The study is limited by the availability of panel data, restricting the geographical scope (147 countries) and time period (1995-2018). While the methodology is robust, future research could benefit from exploring alternative econometric techniques or incorporating additional variables to enhance the analysis. The reliance on trade openness as a proxy for trade protection might also limit the interpretation of findings.