Modern Fiscal Policy and its Impact on Achieving Economic Stability

Abstract

Traditional fiscal policy frameworks face fundamental) structural deficiencies in addressing contemporary) economic shocks, particularly due to their absolute reliance- on retrospective historical data and rigid macroeconomic models. This study aims to present an innovative and unprecedented research model: the construction of a (Digital Twin) based on micro-modeling of virtual agents and reinforcement learning to test the effectiveness of proactive fiscal policies. To achieve this, an advanced software simulation environment was developed using Python and the Mesa library, and the performance of an adaptive (automated finance minister) was compared with that of fixed fiscal policies under severe inflationary shocks. The practical results demonstrated the significant) superiority of smart policy in absorbing shocks and achieving economic stability, reducing recovery time by 30% to 40% and limiting the cumulative erosion of individual wealth. The study concludes that a shift from classical fiscal planning frameworks to interactive computational economics is necessary to ensure the stability of economies in the face of uncertainty.

Keywords: Modern Fiscal Policy, Digital Twin, Agent-Based Modeling, Reinforcement Learning, Economic Stability, Computational Economics.