Green Energy’s Benefits Cross Borders: Global Study Finds Renewables Cut Emissions Far Beyond Their Own Backyard

When a country builds wind farms or installs solar panels, the climate benefits may not stop at its border. A new global analysis of 96 countries over three decades finds that renewable energy consumption reduces carbon dioxide emissions not only within the nation that adopts it, but also—more powerfully—in its neighbours. The study, published in the open-access journal Heliyon, is among the first to separate the domestic and cross-border effects of renewable and fossil fuel energy use on a worldwide scale, and its findings carry a striking implication: the environmental payoff of clean energy investment is largely a shared, regional phenomenon.

The research team, led by Amjad Naveed and Nisar Ahmad together with colleagues from institutions spanning several countries, set out to answer a deceptively simple question with sophisticated tools. Economists have long studied how energy use shapes pollution, but most analyses treat each country as an isolated island. That assumption, the authors argue, is fundamentally wrong. Carbon dioxide released in one nation drifts into shared atmospheric systems, energy markets are linked through cross-border trade, industries relocate across frontiers, and environmental policies diffuse from country to country. Ignoring this geographic interdependence, they warn, produces biased estimates of how energy choices really affect the planet.

To capture these connections, the researchers turned to spatial econometrics, a branch of statistics designed for data that are geographically intertwined. Their framework rested on the Environmental Kuznets Curve, a well-known hypothesis holding that pollution rises during early economic development, peaks at a middle income level, and then declines as economies grow richer, cleaner, and more technologically advanced. The team augmented this classic model with measures of renewable and non-renewable energy consumption and then embedded it in a Spatial Durbin Model, a specification that simultaneously tracks how a country’s own characteristics and those of its neighbours influence its emissions.

The dataset was demanding. To run spatial models, the authors needed a balanced panel, meaning complete annual observations for every country and every variable across the entire study period of 1992 to 2021. After screening the World Bank’s World Development Indicators, which cover more than 200 countries, they assembled a final sample of 96 nations, from major emitters such as China, the United States, and Germany to smaller economies across Africa, Asia, and Latin America. The core variables included carbon dioxide emissions, GDP per capita, the share of renewable energy in total energy use, fossil fuel consumption, and trade openness.

The first step was to test whether spatial dependence actually exists. Using Moran’s I, a standard statistic for spatial autocorrelation, the team found strong and highly significant clustering in every variable of interest. In the panel data, Moran’s I reached 0.789 for carbon dioxide emissions and 0.798 for GDP, both far above the threshold expected under random geographic distribution, with p-values effectively zero. Renewable and non-renewable energy consumption also clustered geographically. In plain terms, countries with high emissions tend to sit near other high-emitting countries, and clean energy adoption tends to cluster regionally as well. This confirmed the study’s first hypothesis and justified the entire spatial modelling exercise.

With dependence established, the researchers compared a family of competing models—ordinary least squares, spatial autoregressive, spatial error, spatial autocorrelation, and spatial Durbin specifications—using likelihood ratio tests to select the best performer. The Spatial Durbin Model won decisively, beating each nested alternative with test statistics that rejected the simpler models at overwhelming significance levels. The model’s spatial parameter, which measures how strongly one country’s emissions respond to those of its neighbours, was positive and significant, confirming that pollution in one nation is statistically intertwined with pollution next door.

The headline results validated the remaining hypotheses. The Environmental Kuznets Curve held: GDP per capita entered positively, its squared term negatively, and both were highly significant, reproducing the classic inverted U-shape in which emissions climb with early growth and fall once economies mature. Notably, when the energy variables were added, the estimated turning point of the curve shifted to a lower income level, suggesting that the energy mix itself accelerates the transition toward environmental improvement. Renewable energy consumption carried a significant negative coefficient, while non-renewable energy consumption carried a significant positive one, and these signs were stable across every model specification tested.

The most provocative findings emerged when the team decomposed the total effects into direct and indirect components, following the standard technique of LeSage and Pace. For renewable energy, the indirect spillover effect accounted for roughly 62 percent of the total emission reduction, dwarfing the 38 percent direct effect. In other words, when a country expands its renewable capacity, neighbouring nations reap more of the benefit than the investing country itself. The authors attribute this to mechanisms well described in environmental economics: clean technologies diffuse across borders through technology transfer, cross-regional investment, and policy harmonisation, so countries that never build a single turbine still see their emissions fall as their region decarbonises.

Fossil fuels told the opposite story. Non-renewable energy consumption raised emissions with a direct effect about twice the size of the spillover component—64 percent domestic versus 36 percent indirect—marking carbon-intensive energy as a predominantly local burden that still leaks across borders. Trade openness, by contrast, showed no significant effect on emissions in the global sample, suggesting that trade alone does not drive cross-border carbon variation unless the underlying energy mix changes. Robustness checks using alternative neighbourhood definitions, including k-nearest-neighbour matrices with k ranging from four to six, preserved every qualitative conclusion, and an alternative fossil fuel measure based on electricity production from oil, gas, and coal yielded consistent results.

For policymakers, the message is unambiguous: renewable energy is a regional public good, and national decarbonisation strategies should be designed accordingly. Because the largest gains from clean energy investment flow across borders, the study argues for coordinated international action—cross-border renewable electricity grids, regional carbon-trading markets, and harmonised policy frameworks—particularly in high-emission regions. The authors also acknowledge limitations: the balanced-panel requirement excluded countries with incomplete data, and unobserved institutional factors may still shape spillover magnitudes. Future work, they suggest, could examine spillovers across groups of countries at different development levels or drill into specific sectors. But the core insight stands: when it comes to clean energy, a country’s best climate strategy may be to help its neighbours go green too.

Subject of Research: Spatial spillover effects of renewable and non-renewable energy consumption on CO2 emissions within the Environmental Kuznets Curve framework

Article Title: Energy and spatial spillovers: How renewable and non-renewable energy impact the environment?

Article References: Naveed, A., Ahmad, N., Zhuparova, A., FathollahZadeh Aghdam, R., Akbar, M., Ali, F., Isatayeva, G., & Beisembayev, G. (2026). Energy and spatial spillovers: How renewable and non-renewable energy impact the environment?. Heliyon, 12(15), Article e45401. https://doi.org/10.1016/j.heliyon.2026.e45401

Image Credits: AI Generated

DOI: 10.1016/j.heliyon.2026.e45401

Keywords: renewable energy, CO2 emissions, spatial econometrics, Environmental Kuznets Curve, spillover effects, fossil fuels, climate policy, Spatial Durbin Model, economic growth, carbon emissions, energy transition, spatial dependence