Adaptive Fiscal Policy Design for Indonesian Coffee Commodities: Transforming Econometric Models to State-Space via Linear Quadratic Tracking Approach

Coffee serves as a strategic commodity for Indonesia's non-oil and gas exports; however, its market dynamics are characterized by high volatility due to global price fluctuations and climate change-induced production uncertainties. Previous research has primarily utilized simultaneous equation models and static optimal control to manage export taxes. A critical limitation of these approaches is their reliance on open-loop strategies, which lack resilience against real-time stochastic disturbances. This study bridges the gap between econometrics and modern control theory by transforming the structural econometric model of the Indonesian coffee market into a reduced state-space form. We propose a Finite-Horizon Linear Quadratic Tracking (LQT) approach to design an adaptive fiscal policy. Unlike static optimization, this method synthesizes a feedback control law that automatically calibrates tax rates in response to market deviations. Simulation results for the 2025–2030 period demonstrate that the LQT-based controller reduces the Sum of Squared Errors (SSE) by 40% compared to traditional open-loop methods and exhibits superior robustness against supply-side shocks. This research provides a novel, robust decision-support tool for policymakers to maintain economic stability under uncertainty.