How Central Bank Independence Shapes Monetary Policy Communication: A Large Language Model Application
Speaker: (opens in a new window)Lauren Leek (European University Institute)
Wednesday, October 30, 14:00–14:45 (Irish time)
Please register (opens in a new window)here to receive the link and password to the online meeting and information on the room at UCD.
Abstract: Although central bank communication is a core monetary policy and accountability tool for central banks, little is known about what shapes it. This paper develops and tests a theory regarding a previously unconsidered variable: central bank independence (CBI). We argue that increases in CBI alter the pressures a central bank faces and amends the reputation costs of not addressing these. We fine-tune and validate a Large Language Model (Google's Gemini) to develop novel monetary policy indices in speeches of 100 central banks from 1997 to 2023. Employing a staggered difference-in-differences and an instrumental variable approach, we find robust evidence that an increase in independence decreases communication portraying central banks to be in full control of their monetary policy conduct and increases communication reflecting financial pressures. We show that these results are not confounded by general changes in communication over time or singular events, in particular, the financial crisis.
About the speaker: (opens in a new window)Lauren Leek is currently a visiting PhD researcher at the London School of Economics, European Institute and a PhD researcher at the Political and Social Science department of the European University Institute in Florence, Italy. She was a PhD trainee at the European Central Bank in the Fiscal Policies division of DG Economics for 6 months. She is interested in applied social and policy questions which can be answered with various quantitative methods. Her interests broadly include: political economy (central banking, taxation, climate, political systems), EU studies (fiscal European integration and democratic reforms), computational social science (text-as-data, large language models, machine learning), and quantitative methods (causal inference, time-series, data science).