Trade-environment nexus under export-oriented and import-driven regimes: Markov regime-switching regression evidence from Uzbekistan

Type of the article: Research ArticleAbstractThis study investigates the nonlinear dynamics of the relationship between foreign trade and carbon dioxide (CO₂) emissions in Uzbekistan over the period 1997–2024. The analysis employs annual time-series data from the World Bank’s World Development Indicators (WDI) database, including three key variables: CO₂ emissions per capita (tonnes), exports of goods and services (current USD), and imports of goods and services (current USD). Using the Markov switching regression (MSR) model, the study identifies two distinct economic states: an export-oriented regime (Regime 1) characterized by high industrial production and export activity, and an import-driven regime (Regime 2) characterized by domestic consumption patterns and elevated import flows.The empirical results demonstrate that the export–emissions relationship is regime-dependent: exports have a statistically significant positive effect on CO₂ emissions only during export-oriented periods (β = 1.54 × 10–10, p < 0.01), while this relationship becomes insignificant during import-driven periods (β = 5.20 × 10–11, p = 0.378). In contrast, imports consistently reduce CO₂ emissions across both regimes (Regime 1: β = −1.05×10–10; Regime 2: β = −1.07 × 10–10, both p < 0.01), indicating a stable import-substitution effect that displaces domestic production-related emissions. The transition probability analysis reveals high persistence in both regimes (P₁₁ = 79.69%, P₂₂ = 81.22%), with structural shifts occurring approximately every five years (expected durations: 4.92 years for Regime 1 and 5.32 years for Regime 2). These findings confirm that the trade–emissions relationship in Uzbekistan is nonlinear and regime-dependent, necessitating the development of regime-sensitive environmental and trade policies.