Evaluating the role of renewable energy investments in CO2E reduction: Analysis through linear, non-linear and ordinary least square model

This study addresses the unresolved question of whether—and under what conditions—renewable energy investment (REI) reduces global CO2 emissions (CO2E). Using a quarterly, world-aggregate panel from 2015 to 2024, we quantify both linear and non-linear REI–CO2E links across seven REI components (wind, solar, geothermal, small hydropower, marine, biofuel, biomass), and test the growth channel through which REI may indirectly affect emissions. Methodologically, we estimate STIRPAT-based specifications via OLS and Fully Modified OLS (FMOLS) for robustness, and complement them with Wavelet Coherence to capture time-varying co-movement. The linear models indicate that a 1 % increase in disaggregated REI is associated with a 0.030–0.048 % rise in CO2E, consistent with construction and supply-chain emissions. Non-linear estimates uncover investment thresholds: an inverted-U for solar and wind, with CO2E declining only beyond log thresholds of 29.032 (≈$1.82 trillion, solar) and 27.107 (≈$267.5 billion, wind); and a U-shape for small hydropower and biofuel, where CO2E increases beyond 21.190 (≈$522.7 million) and 23.400 (≈$8.071 billion), respectively. Interaction terms show that solar, wind, and geothermal investments reduce CO2E via the economic-growth channel (negative REI × GDP effects), whereas biofuel investment raises CO2E as growth scales. Control variables behave as expected: population and trade openness increase CO2E, while higher income per capita is associated with lower emissions over the study window. Wavelet evidence confirms time-scale heterogeneity in REI–CO2E co-movement. prioritize large-scale solar and wind to surpass effective thresholds while applying strict sustainability screens to biofuel and small hydro, supported by targeted R&D, concessional finance, and efficiency-oriented carbon pricing.