Descripción del proyecto
Policy decisions about government borrowing, taxation, and spending have a differential impact on generations and thus imply substantial intergenerational transfers. The costs and benefits of these policies occur over decades and come with large uncertainty, for instance related to productivity growth, global financial conditions, and demographic trends. Therefore, fiscal policies, whether intentional or not, transfer risk and resources across generations. Assessing their impact on welfare calls for simulating such policies in models that include the relevant demographics and risks, yet also the related risk-premia. The latter lower government borrowing costs relative to the expected return on investments and are thus particularly important for the welfare implications of fiscal policy. SOLG for Policy aims to build stochastic overlapping-generations (SOLG) models that integrate all these elements and are thus able to quantitatively assess the welfare implications of fiscal policy choices involving intergenerational redistribution and risk-sharing. To solve these models despite their high-dimensional nature, SOLG for Policy leverages recent advances in computational economics based on adaptive sparse grids (ASGs) and develops three extensions of this approach. First, using shape-preserving ASGs to facilitate value function iteration. Second, merging ASGs with endogenous grids to avoid compute-intensive optimization steps. Third, parameterizing intergenerational distributions to further alleviate the curse of dimensionality. Policy questions that will be addressed include: Should the debt accumulated in the Covid crisis be paid down through taxation or do low real interest rates make it desirable to leave the debt and its uncertain burden for future taxpayers to deal with? Is it feasible and desirable to make the pension system provide less redistribution yet more insurance across generations, and how can this be achieved?