Jon Faust Mini-Course Descriptions

Jon Faust will give two presentations on the topic of identifying causal structure in macroeconomic data. Causal inference attempts to go beyond describing what happened in a dataset to explaining why things happened the way they did. While mere data summary may be sufficient for many important questions (e.g., has there been a great moderation in the volatility of U.S. GDP?), we often want to know more: what role did globalization, or monetary policy, or improved inventory management play in precipitating the great moderation?
 
The presentations are intended as a self-contained course, taking issues from beginning to end. The course will be at an advanced level, however, and will be most accessible to those with a basic familiarity with classical issues regarding identification in linear models (e.g., rank and order conditions and instrumental variables).

In the first lecture, I will start with basic conceptual issues in causal inference and lay out a framework for organizing our thinking about causality. Then we will apply this framework to analyze examples of causal inference using vector autoregressions, generalized method of moments estimation, and dynamic stochastic general equilibrium models. By studying successes and failures in applied work within a unified conceptual framework, we will arrive at some practical tips for generating more productive research on why the macroeconomy behaves the way it does.