SMU SOE Seminar Series (September 4, 2026): Tracing Data Through Analytical General Equilibrium

TOPIC:

ABSTRACT

We develop a heterogeneous-household framework with closed-form individual and aggregate dynamics jointly capturing growth, business cycles, and distributional change. Three moments (aggregate capital and two wealth-sorting covariances) suffice for the equilibrium, which is invariant to household forecasts and transversality. The model is exactly invertible at both macro and micro levels: it reads observed data as an equilibrium path and recovers the primitives, prices, and distributional states consistent with it. In U.S. aggregates (1950–2019), wealth concentration emerges from TFP-driven return-wealth sorting, and feeding the recovered shocks forward forecasts output and hours more accurately than a reduced-form time-series model. At the household level, inverting PSID and SCF microdata yields the joint distribution of individual parameters, which reproduces observed life-cycle and cross-sectional patterns (a hump-shaped age profile of productivity, sharp wealth sorting) and agrees across the two surveys. Counterfactuals on these households show that persistent capital-tax changes produce decades-long distributional drift, with the gains from growth accruing disproportionately to the wealthiest.