Topic #2: Novel instruments/proxies for structural shock identification

Outline.

Shock instruments or proxies in macroeconomics are designed to isolate the effects of a specific structural shock of interest on the macroeconomy. The instrument/proxy should be highly correlated with this specific shock but truly exogenous to other factors and structural shocks. The macroeconomic literature generally agrees on how to effectively measure monetary policy surprises, extreme weather, or climate shocks in an exogenous manner by assessing the high-frequency responsiveness of financial markets (interest rates at different maturities, prices on futures markets, stock returns, exchange rates, etc.) around policy announcements or significant events/phenomena. A dissertation related to this topic could focus on designing/finding a novel and innovative instrument to identify a particular structural shock of interest (e.g., extreme weather or climate shocks) and analyse its features using an event study. Instruments can also be constructed from text. For example, Aruoba and Drechsel (forthcoming, AEJ: Macro) use natural language processing of documents prepared by Federal Reserve staff to identify monetary policy shocks. Along these lines, you may want to use a large language model (LLM) to construct an instrument (e.g., for OPEC announcements).

VoxEU columns.

  • What we know about climate change and inflation. [VoxEU]
  • Global weather disruptions, food commodity prices, and economic activity: A global warning. [VoxEU]

Suggested initial readings.

  • For examples of how instruments are constructed, see Känzig (2021) for high-frequency surprises around OPEC announcements, Alessandri & Gazzani (2025) for an instrument based on daily news on the gas market, and the weather and climate papers in the list of available shock series and instruments.
  • Aruoba, S. B. & Drechsel, T. (forthcoming). Identifying monetary policy shocks: A natural language approach. American Economic Journal: Macroeconomics. [DOI]