Carbon Intensity, Debt Concern, and Green Growth
Overview
This completed empirical report evaluates whether fiscal pressure changes the relationship between carbon intensity and green-growth performance. The portfolio presents the stable baseline result and preserves the divergence in the robustness models.
Research Question
Does debt concern amplify the adverse conditional association between carbon intensity and green growth?
Economic or Technical Motivation
Debt pressure can restrict fiscal space for transition investment and make carbon-intensive production harder to replace. An interaction model tests whether the carbon-intensity slope varies with debt concern.
Data
The baseline panel contains 552 observations for 69 countries across eight periods. The IV and dynamic specifications use 483 observations because of lag and instrument requirements. Controls include regulatory quality, trade openness, and human development.
Methodology
The baseline specification uses country and year fixed effects with clustered standard errors. The report also estimates IV fixed effects and a dynamic panel, and reports Hansen and AR(2) diagnostics for the latter.
Main Findings
The baseline fixed-effects model reports carbon intensity at -1.568, debt concern at -0.638, and the interaction at -0.301; each is significant at the reported 5% level. The within R-squared is 0.294. These coefficients support a negative conditional association in the baseline model only.
Robustness, Validation, or Model Assessment
The IV interaction is +2.804 with a standard error of 4.491, while the dynamic interaction is -0.823 with a standard error of 0.911. Neither reproduces the baseline interaction precisely. Hansen p-values are 0.719 and 0.554 in the reported dynamic variants, and AR(2) p is 0.375. The robustness evidence therefore limits any causal interpretation.
Tools and Technologies
Stata, country fixed effects, year fixed effects, clustered inference, interaction models, IV fixed effects, and dynamic-panel diagnostics are directly evidenced.
Limitations
The preferred baseline relationship is not stable across the IV and dynamic specifications. Measurement choices for debt concern and green growth may also affect interpretation. No causal claim is made.
Deliverables
- Completed regression report
- Fixed-effects, IV, and dynamic-panel tables
- Stata code and model diagnostics
- Marginal-effect interpretation