New Keynesian Models, Puzzles and Omitted Effects

The modern macroeconomic debate grapples with explaining complex phenomena, often giving rise to theoretical and empirical puzzles. In response, New Keynesian models have evolved, incorporating frictions and agent heterogeneity, leading to the development of TANK and HANK frameworks. However, increasing model complexity does not always clarify underlying causes. This work introduces a New Keynesian model with Omitted Effects, incorporating previously unaccounted-for variables into agents’ objective functions. Similar to Omitted Variable Bias in econometrics, neglecting key variables in a general equilibrium setting leads to model misspecification and puzzling anomalies. The study follows both theoretical and empirical approaches: first, demonstrating how Omitted Effects address the Forward Guidance Puzzle and the consumption response to government spending; second, using an algorithm to identify omitted variables that improve model fit to US data. Empirical findings highlight the crucial role of real wages, inflation, and income Omitted Effects in resolving key macroeconomic inconsistencies. This framework enhances DSGE models’ flexibility, providing a systematic tool to refine economic modeling and deepen the understanding of macroeconomic dynamics.