Paradise lost? A brief history of DSGE macroeconomics

Since the Global Financial Crisis, academic economists and policymakers have had to deal with uncomfortable questions about the quality of their models and the state of macroeconomics as a profession. This note offers a summary of this discussion, focusing on the Dynamic Stochastic General Equilibrium (DSGE) framework and its underpinnings. This class of models reflects both theoretical advances and perennial modeling challenges. While DSGE modeling developed in times of scarce micro data and limited computational resources, it has much room for improvement given progress along these dimensions and advances in other branches of economics. Key tasks on the to-do-list for model improvement include the modeling on the financial sector, departures from the representative agent and rationality, as well as clarification of the empirical relevance of the Lucas critique. The framework is likely to remain a major research and policy tool, although its limitations call for greater robustness, validation and open recognition of uncertainty in drawing real-life quantitative conclusions.