Can Renewable Energy Really Extend Human Life? Study of 15 Leading Economies Challenges the Clean Power–Longevity Link

The global shift toward renewable electricity promises more than cleaner energy systems. By reducing dependence on fossil-fuel combustion, the transition is expected to lower harmful pollution and contribute to healthier, longer lives. However, a fundamental question remains unresolved: does expanding renewable electricity continue to deliver measurable longevity gains in countries where clean power already accounts for a substantial share of generation?

A study titled "Is Non-Hydro Renewable Electricity Associated with Longer Life Expectancy? Quantile and Machine-Learning Evidence from Fifteen Leading Adopter Economies, 1990–2021," published in the journal Energies, challenges expectations of a straightforward relationship. Conducted by Avazbek Sadikov, Dostonbek Eshpulatov and Raufhon Salahodjaev, the research examines 15 leading renewable electricity adopters over 32 years, combining advanced econometric techniques with machine-learning analysis.

Higher renewable electricity penetration was not significantly associated with longer life expectancy in the principal statistical models. The researchers found no convincing evidence that the relationship strengthened among higher-longevity observations or after renewable electricity reached particular thresholds. However, the absence of a detectable association does not establish that renewable energy lacks health benefits.

Clean electricity's health promise faces a tougher test

Replacing fossil-fuel electricity generation with renewable alternatives is widely expected to reduce exposure to harmful pollutants. Coal- and oil-based generation produces fine particulate matter, sulfur dioxide, nitrogen oxides and other substances linked to cardiovascular and respiratory diseases. Expanding wind, solar and other renewable technologies can reduce these emissions when they displace fossil-fuel generation.

Earlier international research has frequently identified positive relationships between renewable energy consumption and life expectancy. Studies covering large groups of countries and major advanced economies have associated cleaner energy systems with improved longevity, although results vary across income groups, technologies and methodological approaches.

However, much of the existing evidence concerns economies in the early or intermediate stages of renewable adoption. During those phases, replacing heavily polluting electricity generation may offer substantial environmental advantages. Whether additional renewable capacity delivers comparable health returns after electricity systems have already undergone significant transformation remains less certain.

The researchers addressed this gap by examining Denmark, El Salvador, Nicaragua, Uruguay, Kenya, Lithuania, Luxembourg, Estonia, Portugal, the United Kingdom, Germany, Spain, Ireland, the Netherlands and Finland. Their selection reflects the highest qualifying shares of non-hydroelectric renewable electricity in 2021, with complete data available from 1990 onward.

The sample encompasses 480 country-year observations and considerable economic diversity. Renewable electricity shares in 2021 ranged from 31% in Finland to 78.9% in Denmark. Its inclusion of European economies alongside renewable electricity pioneers in Africa and Latin America provides a basis for examining countries with different development circumstances.

Rather than relying on simple comparisons, the study assessed renewable electricity penetration alongside income per person, urbanization and carbon dioxide emissions. Statistical techniques accounted for country-specific characteristics and common international trends, while additional models examined whether the relationship differed across the life-expectancy distribution.

More renewable power, but no measurable longevity dividend

The principal statistical analysis found no significant relationship between annual changes in non-hydroelectric renewable electricity shares and changes in life expectancy. Its estimated coefficient was positive but small, at 0.000128, with a p-value of 0.379, providing insufficient statistical evidence for a reliable association.

The result complicates interpretations that treat renewable electricity expansion as a direct predictor of national longevity. Although cleaner generation can reduce pollution under appropriate conditions, life expectancy also reflects healthcare, income, sanitation, education and demographic developments. The study could not isolate a measurable longevity contribution from renewable penetration within its selected economies.

Researchers also examined whether the association differed across the conditional distribution of life expectancy. Their quantile estimates remained small and statistically insignificant throughout, ranging from 0.000141 at the lower end to 0.000117 at the upper end. Formal testing identified no meaningful difference between the extremes.

A separate investigation considered whether renewable electricity penetration needed to exceed a particular threshold before a stronger association became visible. Of the 480 observations, only 46 represented renewable shares exceeding 40%, limiting the evidence available for examining highly renewable-dependent electricity systems.

The researchers found no statistically significant change in the relationship above that threshold. A broader search across alternative penetration levels also produced unstable results, with estimated thresholds varying considerably across statistical replications. Consequently, the study could not identify a reliable point beyond which renewable electricity expansion became more strongly associated with longevity.

These results do not demonstrate diminishing health benefits or establish that advanced renewable adopters have exhausted potential gains. They indicate that neither a consistent positive association nor a distinct high-penetration threshold could be identified using the available national data and methods.

A post-Soviet mortality crisis exposes a hidden statistical trap

One of the study's most revealing findings concerns the danger of interpreting statistically significant results without examining the historical events behind them. Certain models initially suggested a negative relationship between earlier renewable electricity penetration and subsequent life expectancy, potentially implying that cleaner electricity was associated with shorter lives.

The researchers traced this counterintuitive pattern to the post-Soviet mortality crisis affecting Lithuania and Estonia during the early 1990s. Lithuanian life expectancy declined from 71.2 years in 1990 to 68.5 years in 1994, while Estonia experienced a decrease from 69.5 to 66.5 years.

Both economies subsequently experienced substantial expansion in non-hydroelectric renewable electricity generation. Their demographic histories therefore combined an earlier mortality shock with later renewable growth, creating a sequence that some statistical models could misinterpret as a delayed negative energy–health relationship.

When Lithuania and Estonia were excluded, the significant negative association disappeared. Excluding the 1990s produced a similar disappearance of statistical significance. Although these tests cannot conclusively establish the underlying mechanism, they strongly suggest that unrelated demographic developments influenced the original estimates.

The analysis also revealed sensitivity to country selection. Restricting the dataset to the ten leading renewable adopters produced a significant negative coefficient, while removing individual countries sometimes changed the estimated direction of the relationship. Such instability reduces confidence in interpreting any isolated estimate as evidence of an underlying energy-related health effect.

The researchers consequently emphasize the importance of historical context, particularly when analyzing countries affected by major demographic disruptions. Statistical associations can reflect overlapping national trajectories rather than meaningful relationships between the variables under investigation.

Machine learning strengthens the case for cautious energy-health claims

To investigate possible relationships that conventional statistical models might overlook, the researchers compared their econometric analysis with machine-learning approaches, including XGBoost and random forests. These models can identify complex patterns without requiring every relationship to follow a predetermined linear structure.

The machine-learning results offered little additional support for a substantial renewable electricity–longevity association. Income per person emerged as the strongest measured predictor, followed by the common time trend, urbanization and carbon dioxide emissions. Renewable electricity penetration contributed the smallest average predictive attribution among the principal variables.

More sophisticated models also failed to outperform a simple forecasting benchmark. Predicting that a country's life expectancy would equal its previous year's value produced lower out-of-time forecasting errors than either XGBoost or random forests. The finding demonstrates how strongly historical persistence influences longevity predictions.

Although machine-learning models identified patterns in the data, their predictive contributions could not establish causal relationships. The researchers caution that algorithms may capture differences in national development, economic structures and long-term trends without identifying the independent health contribution of renewable electricity.

Several limitations further qualify the conclusions. The analysis covers only 15 economies, reducing its ability to detect modest associations. Country selection depends on renewable penetration at the end of the study period, while national averages conceal differences in local pollution exposure and population health.

Crucially, the study did not directly measure particulate pollution, fossil-fuel displacement or individual exposure to emissions. Carbon dioxide emissions served as an explanatory variable, but they cannot represent the local pollutants through which electricity generation affects respiratory and cardiovascular health. Important determinants such as healthcare expenditure, sanitation, education and energy access were also absent from the statistical models.

For governments and development institutions, the findings therefore call for greater precision when evaluating renewable electricity investments. Health benefits should be assessed using evidence of actual pollution reductions, exposure changes and mortality outcomes, rather than inferred automatically from increases in renewable electricity shares.

The research offers no basis for concluding that renewable deployment damages health or that further investment lacks environmental value. Instead, it establishes a narrower but important distinction: the anticipated public-health advantages of cleaner electricity are not necessarily detectable through changes in aggregate life expectancy.

Future investigations will need more direct environmental and health measurements, stronger causal research designs and closer attention to differences between electricity systems. Until such evidence becomes available, renewable energy's contribution to human longevity remains a plausible but insufficiently quantified benefit in the economies examined.