Oil Price Shocks and CO2 Emissions in Azerbaijan: Evidence from a Multiple Threshold Nonlinear ARDL Approach

Abstract

This study investigates the regime-dependent effect of Brent petroleum price shocks on CO₂ emissions in Azerbaijan for the period of 1990-2024. Azerbaijan is an oil-exporting, resource-dependent economy, so changes in Brent prices may influence emissions through oil revenues, fiscal expansion, investment activity, energy use and energy-intensive production. Unlike prior research that investigates factors affecting CO₂ emissions and the macroeconomic impacts of oil prices separately, this paper directly analyzes the nexus between Brent oil price shocks and CO₂ emissions within a Multiple Threshold Nonlinear ARDL framework. First, Harvey-type linearity tests show that LCO₂ and LBPP are nonlinear. The unit root tests, such as ADF, Flexible Fourier ADF and Fractional Flexible Fourier ADF tests, reveal that the variables are integrated of order one and are suitable for ARDL-type modeling. The results of MTNARDL confirm the long-run relationship between Brent price shocks and CO₂ emissions. The error correction coefficient is significant and negative, implying a rapid adjustment towards the long-run equilibrium. The effects in the long-run are mostly observed in the Q1 and Q2 regimes, while the effects in the short-run are concentrated in the Q3 and Q4 regimes. The strongest long-run response is observed in the Q2 regime, suggesting that weak or transitional changes in Brent prices are particularly relevant to the long-run behavior of CO₂ emissions. Further Wald asymmetry tests corroborate that the relationship is regime-dependent and not linear and uniform. The results show that the environmental consequences of oil price shocks in Azerbaijan depend on both the size of the shock and the time horizon considered.