Quantile-based assessment of energy-CO2 emission nexus in Pakistan

Abstract

Nonrenewable energy sources maintain a substantial majority of Pakistan’s energy composition, exceeding 70%, posing challenges to achieving sustainability goals for a low-carbon economy. Recognizing this, the study determines the critical thresholds where renewable and nonrenewable energies affect more significantly on CO2 emission over the period from 1972Q1 to 2020Q4. The analysis begins by confirming the stationarity of the data through quantile unit root analysis, followed by an examination of long-term associations using quantile cointegration. For quantile-based impact assessments, we apply quantile regression. To uncover the direction of causality within quantiles, we use a novel approach, quantile causality analysis. Nonrenewable energy sources exhibit a long-term association at disaggregated levels, whereas the same is not true for renewable energy across the quantile distribution. Quantile regression results reveal that renewable energy sources positively impact CO2 emissions, with coal having the highest coefficient, followed by oil and gas, particularly in the upper quantiles, τ = {0.70–0.75}. However, renewable energy sources prove insignificant in decreasing CO2 emissions. Similarly, total energy consumption has a positive influence on CO2 emissions at extremely low quantiles, τ = {0.05–0.30} and high quantiles, τ = {0.65–0.90}, indicating sensitivity to extreme variations. The quantile causality analysis highlights a bidirectional causality relationship among CO2 emissions, total energy consumption, and both renewable and nonrenewable energy consumption across lower and upper quantiles. Policymakers should reallocate funds, provide subsidies, and introduce infrastructure development projects to reduce the burden on nonrenewable energy sources based on this quantile analysis.