Optimal Allocation of Carbon Tax Revenue in the State Budget: A Linear Programming Approach with Statutory Earmarking Constraints

Moch Panji Agung Saputra (Department of Mathematics, Faculty of Mathematics and Natural Sciences, Universitas Padjadjaran, Sumedang, Indonesia)
Cece Mulyadi (Sociology Department, Faculty of Social and Political Science, Universitas Padjadjaran, Sumedang, Indonesia)
Wandi Adiansah (Department of Social Welfare, Faculty of Social and Political Science, Universitas Padjadjaran, Sumedang, Indonesia)

International Journal of Business, Economics, and Social Development
Vol. 7 No. 2 (2026): International Journal of Business, Economics, and Social Development (IJBESD)

The enactment of Law No. 7/2021 on the Harmonization of Tax Regulations (UU HPP) has mandated Indonesia's first carbon tax regime, with revenues designated for reallocation within the state budget (APBN). However, no mathematical framework currently exists to optimally determine how these revenues should be distributed across competing statutory priorities, namely, social protection transfers, renewable energy subsidies, green infrastructure investment, and fiscal buffer reserves, while simultaneously satisfying legally binding earmarking constraints. This paper addresses that gap by proposing a Linear Programming (LP) model for the optimal allocation of carbon tax revenue under APBN statutory constraints. Given projected revenue as a deterministic function of the carbon tax rate, the LP model maximizes an aggregate social-environmental benefit objective subject to three statutory earmarking bounds: a minimum social protection floor (α = 35%), a minimum combined green expenditure floor (β = 40%), and a maximum fiscal buffer ceiling (γ = 15%). The model is calibrated using empirical APBN parameters and evaluated across three statutory tax rate scenarios: IDR 30,000/ton CO₂e (statutory minimum), IDR 82,500/ton (policy-optimal midpoint), and IDR 150,000/ton (upper bound). Results are further compared across three earmarking policy regimes (Pro-Social, Balanced, and Pro-Environment) to generate a decision matrix for fiscal authorities. The analysis demonstrates that the LP framework consistently produces feasible, globally optimal allocations across all scenarios, with the Policy-Optimal scenario generating IDR 70.1 Trillion in annual revenue and a fully statutory-compliant allocation. Sensitivity analysis of the earmarking parameters reveals that marginal relaxation of the fiscal buffer ceiling yields the largest efficiency gain of the three constraints. The findings provide directly actionable guidance for Indonesia's Ministry of Finance in designing carbon revenue recycling policy under UU HPP.