A Game Theory Approach Simulating Integrity in Kenya's Carbon Market

The Python script developed for this project implements a tripartite evolutionary game model to simulate strategic behavior in Kenya's carbon credit market. Built around replicator dynamics — a system of ordinary differential equations (ODEs) — it models how three key actors (Environmental Regulators, Power Generating Entities, and Third-Party Verification Agencies) adjust their strategies over time in response to enforcement pressure, market payoffs, and reputational incentives. The script uses scipy.integrate.solve_ivp to numerically integrate the ODE system across four policy scenarios (voluntary, compliance, hybrid, and circular economy), computes terminal equilibria to identify Evolutionarily Stable Strategies (ESS), and generates trajectory plots and sensitivity analyses via Matplotlib by sweeping key parameters such as penalty probability and detection rates. It is structured as a modular package with dedicated functions for defining payoffs, running simulations, and stress-testing across multiple initial conditions, and includes a pytest test suite to validate convergence behavior under each scenario.

Carbon credits have emerged as a market-based solution to the global challenge of climate change. Although formalized through international treaties and heavily regulated, integrity concerns over their quality and veracity increasingly underline their trade, especially in the Global South. This study seeks to simulate how regulatory strategies, compliance behavior and verification integrity react to the influence of enforcement pressure, reputational cost and market incentives. This study adopts the game theory model approach and uses a Python script to model actor strategies, payoff parameters, and replicator dynamics in strategic interactions between power producing entities, third party verification agencies, and environmental regulators in Kenya. Findings show that when enforcement is minimal and penalties are weak, there is rapid convergence towards deceptive behavior and strict regulation stabilizes when the cost of enforcement is offset by reputational gains and the deterrent effect of penalties. Also, when community actors are empowered to monitor and report, their influence modifies actor payoffs and reduces the attractiveness of collusion. More importantly, the study finds that the minimum penalty of three times the collusion payoff is required to deter deceptive behavior and collusion. The study concludes with recommendations to strengthen market integrity such as addressing verifier bias, making penalties proportional to misconduct, addressing information gaps and compulsory inclusion of communities in project management.