Impact of Economic Drivers and Climate Change Policies on the Stability of French Pension System

In this paper, I present a machine learning (ML) model, trained using the dataset pre-sented by the “Conseil d’Orientation des Retraites” (COR), that simulates pension ex-penditures for France for the period 2025-2070 for 12 scenarios. Six of these scenarios are presented for the first time in this study, and another six were proposed by the Network for Greening the Financial System (NGFS) initiative. The main result of this study in-dicates that Life Expectancy at 65 (LE@65) is not the most significant variable for planning pension expenditures over the next decades. LE@65 is only the most important ex-penditure driver under some conditions (e.g. productivity < 0.7%). For higher values of productivity, LE@65 is only a second-order variable, which is superseded by others, such as GDP and then social costs. Additionally, I show that GDP growth is crucial to the system stability, while the type of production and inflation trends also play a significant role. At last, predictions based on the NGFS scenarios suggest that climate change policies will have a substantial impact on the French pension system through a series of shocks on GDP until 2050.