Econometrics for Macroeconomic Policy Evaluation

Ziel

Traditionally, the evaluation of macroeconomic policy has relied on analyzing specific economic models which often do not correspond to the complex and data-rich environments in which real world policy decisions are made.

To this extent, I propose to develop a new framework for evaluating macroeconomic policy decisions ---policymetrics--- that is inspired by the robust inference literature in econometrics and makes minimal assumptions on the underlying economic model. The methodology detects optimization failures, i.e. instances in which policy decisions do not minimize the (welfare) loss function, and determines the causes and dynamics of such failures, all while fully respecting the limitations imposed by the Lucas critique.

Part 1. The detection of optimization failures is based on the gradient of the loss function evaluated at the proposed policy choice. The key insight is that for large classes of macroeconomic loss functions the gradient only depends on known or estimable sufficient statistics such as forecasts, dynamic causal effects and preferences. I will tailor the methodology to allow for constrained policy problems, risk aversion and the real-time detection of optimization failures.

Part 2. The methodology for understanding optimization failures is inspired by the forecast evaluation literature and aims to infer which trade-offs, across mandates and horizons, dynamics and variables were overlooked when setting policy. Among others, I will develop methods that test for systematic optimization failures, optimization encompassing and breaks in policy performance.

I will apply the framework to evaluate contemporaneous and historical monetary, fiscal and climate policy decisions from different governments and institutions around the world, hereby providing a unique non-model based assessment of modern macroeconomic policy.

Wissenschaftliches Gebiet (EuroSciVoc)

Dieses Projekt wurde noch nicht bei EuroSciVoc klassifiziert.
Schlagen Sie die Wissenschaftsbereiche vor, die Ihrer Einschätzung nach besonders relevant sind, und helfen Sie uns, unseren Klassifizierungsdienst zu verbessern.