This study examined how carbon pricing and compliance costs affect corporate trading behaviour in carbon markets. It addressed a gap in understanding how market-oriented climate policies shape corporate conduct at the micro level. The study used a two-stage quantitative methodology and analysed 5,000 firm-level records from four sectors (cement, energy, manufacturing, steel) and multiple fuel types. Multiple linear regression identified determinants of carbon prices and compliance costs. Logistic regression assessed whether these factors could predict firms’ trading decisions—whether to purchase or sell allowances. Sensitivity and scenario analyses tested model robustness under theoretical conditions: high emissions, policy tightening, demand shock, and price escalation. Results showed very weak links between operational variables and market outcomes. The carbon price regression had an R² of 0.0019, and the compliance cost regression had an R² of 0.0007, indicating that firm-level factors explain less than 0.2% of price variation. This is not a statistical failure—it’s the key finding and the novel contribution of this study: treating near-zero explanatory power as substantive evidence that carbon pricing and trading are driven mainly by external policy frameworks, regulatory uncertainty, and strategic anticipation—not current emissions or compliance costs. The logistic regression’s accuracy was 0.50 and ROC-AUC 0.48, consistent with random classification, further supporting the conclusion. The results show that effective carbon markets need more than price signals. Clear regulatory rules, stable allowance mechanisms, and supportive institutional infrastructure are crucial for aligning firm incentives with real emission-reduction objectives.