Renewables do not cut carbon emissions equally across countries or technologies. Analysing six energy sources in 27 OECD countries, Gaetano Perone finds that geothermal/biomass and hydropower deliver the strongest long-run reductions, followed by solar, while wind and biofuels have smaller effects. The findings highlight why policymakers should prioritise renewable sources that match national resources and actual emissions performance.
As governments increase investments in renewable energy, a key policy concern remains: does their chosen method cut carbon dioxide (CO2) emissions in a statistically significant and causally relevant way? Using fresh data from 27 OECD countries over the past six decades, my research finds that the relationship between renewable energy and CO2 emissions is more complex than previously thought, and heavily influenced by country-specific factors as well as the type of renewable energy used.
Where renewables can work best: lessons from OECD countries
The empirical analysis in the paper is based on aggregated panel data from 27 countries within the Organisation of Economic Co-operation and Development (OECD) countries and spans the period from 1965 to 2020. The study found that renewable energy production significantly reduces CO2 emissions over time. Among the six energy sources evaluated, geothermal and biomass (GEOB) and hydropower have the biggest impact on lowering emissions. Solar energy also makes a significant contribution to decarbonisation, although the weather conditions and the initial investment deficit in several countries limit its effectiveness. Wind energy and biofuels also reduce CO2 emissions, but their effects are more modest. For wind, this likely reflects lifecycle emissions from turbine manufacturing, installation, and transport. For biofuels, it reflects trade-offs between land use and energy production.
My research found that, for every 10 terawatt-hours (TWh) increase in renewable energy production, GEOB reduces emissions by approximately 1.17 metric tons per capita, hydropower by 0.87, and solar by 0.77 metric tons. Wind and biofuel have a smaller but still negative influence, with decreases of approximately 0.21 and 0.19 metric tons per capita.
The research also showed that nuclear energy had no statistically significant effect on long-term emissions. This could represent the varied usage of nuclear across OECD countries, as well as the long-term environmental and financial concerns involved with nuclear development.
My Granger causality analysis reveals a relationship between CO2 emissions and renewable energy production. Hydropower, solar, and wind have a unidirectional effect on CO₂ reductions, whereas GEOB and biofuel show a bidirectional relationship. In simple terms, this means that some renewables appear to help drive CO2 reductions, while for GEOB and biofuels, the relationship works in both directions: renewable production and emissions seem to influence each other over time. The analysis demonstrates that boosting renewable energy, particularly solar, hydropower and geothermal energy, is an effective approach to reducing emissions in OECD countries. Figure 1 summarises the heterogeneity across energy sources, showing both the estimated reduction in CO2 emissions for each additional 10 TWh of production and the direction of the causal relationship.
Figure 1. Estimated impact of renewable energy production on CO2 emissions.
What does this mean for policymakers?
While the econometric estimates are based on an aggregate panel of 27 OECD countries, structural characteristics such as per capita electricity generation by source, natural resource endowments, and potential for expansion vary widely across countries. The heatmap below (Figure 2) illustrates the average annual electricity generation per million inhabitants across five renewable sources and nuclear from 2010 to 2020.
Figure 2. The average annual electricity generation per million inhabitants from 2010 to 2020 (in TWh, annotated with original values).
These differences suggest differentiated but complementary policy priorities.
Canada, Iceland and Norway, which have long coastlines and cold, dark climates, exhibit the highest per capita electricity generation from hydropower, confirming their natural advantage in this sector. This is consistent with their high latitude, lower solar potential, and large water resources, which make hydropower a more natural fit than solar energy. As could be expected, solar generation remains negligible in line with their low long-term solar potential (PVOUT). These countries should focus on reinforcing existing hydropower systems and simplifying licensing procedures for small-scale, decentralised plants.
Chile, Mexico and Spain, which all enjoy significant sunshine, report relatively low per capita solar generation. Given this underutilised photovoltaic capacity, fiscal and regulatory measures, such as feed-in tariffs, tax incentives, and streamlined permitting, could substantially accelerate the adoption of solar energy, particularly in high-radiation regions.
Italy and Greece demonstrate relatively high per-capita solar generation (despite only moderate PVOUT), moderate wind penetration, and limited output from geothermal and biomass sources. Policies that mitigate the financial risks of early-stage geothermal exploration, such as public co-financing or risk guarantees, could unlock greater diversification in their renewable energy mix.
Germany, having recently phased out nuclear, exhibits comparatively high per-capita generation from both solar and wind sources. Its primary challenge lies in managing intermittency. Strategic investment in next-generation grid infrastructure and cross-border interconnections should be prioritised to support system integration and energy security.
Cyprus, Israel and Poland remain among the lowest performers in overall renewable output. These countries may benefit from prioritising low-cost technologies such as onshore wind or rooftop PV, depending on land availability and administrative capacity. High PVOUT in Cyprus and Israel suggests that scaling up solar generation should be a central policy objective supported by accelerated permitting and simplified procedures.
America records average per capita output from solar and wind energy, as well as high biofuel generation. A promising strategy would be to further develop hydropower, given its relatively high technical potential compared to Europe and the generally low cost of hydroelectricity.
France and South Korea rely more heavily on nuclear energy. While nuclear output was not found to be significantly associated with per capita CO2 reductions in the econometric model, it continues to play a key role in baseload electricity generation. Future policy directions should carefully weigh long-term cost-effectiveness, public acceptability, and energy security concerns.
Australia shows moderate per-capita generation from hydropower, solar and wind with minimal contribution from other renewable sources. Ample land availability and high PVOUT support the expansion of battery storage and flexible grid solutions, as well as opportunities to strengthen hydropower, thereby improving system resilience.
Finally, Sweden presents one of the most balanced and diversified renewable portfolios, with high per capita output from hydropower, wind, nuclear, and geothermal sources. Sustaining this trajectory will require continued grid modernisation and credible long-term investment frameworks to meet carbon neutrality goals.
These results are especially relevant to the Green Deal in Europe
Simply increasing renewable energy capacity will not be sufficient to satisfy climate targets. Achieving net-zero emissions requires not only investment in renewable energy but also the successful integration of these sources into energy networks. This entails modernising the grid, aligning regulations, and developing consistent policy frameworks to ensure a seamless transition. To adequately assess member-state achievement, the EU should shift its focus from capacity-based measurements to performance-based criteria that capture actual emission reductions and system-wide efficiency. Renewables are crucial, but they must be part of a larger structural transition to enable long-term decarbonisation.
*All data used in this analysis are publicly available and included in the published article.
This blog is based on The relationship between renewable energy production and CO2 emissions in 27 OECD countries: A panel cointegration and Granger non-causality approach, published in Journal of Cleaner Production (2024).
This article gives the views of the author, not the position of LSE Business Review or the London School of Economics. You are agreeing with our comment policy when you leave a comment.
Image credit: Lee Charlie provided by Shutterstock.