Large Language Models – Bank Underground

Dragos Gorduza, Livia Silva Paranhos and Ivan Yotzov

Monetary policy decisions reflect incoming data, forecasts, and policymaker preferences. In this post, we use large language models (LLMs) to analyse Bank of England Monetary Policy Committee (MPC) transcripts from 2015 to 2017. We extract structured measures of individual views on inflation and growth, and the narratives justifying policy decisions. This gives a new perspective on UK monetary policy makers’ reasoning, using recently published transcripts. We show that concerns about economic growth relative to its potential weakened and later strengthened over the sample, while concerns about above-target inflation increased. We link these measures to voting behaviour and show that, over the sample period, both growth and inflation measures predict votes for higher rates, but only growth measures explain votes to cut Bank Rate.

Continue reading →

Iulia Bucur and Ed Hill

Modern language models – think OpenAI’s GPTs, Google’s Gemini or DeepSeek – are powerful tools: but how can we use them in economic policymaking? Economic analysis often relies on decompositions to understand macroeconomic data and inform counterfactuals. But these decompositions are typically obtained from numerical data or macroeconomic models and so may overlook nuanced insights embedded in unstructured text. We propose decomposing the metrics which Large Language Models (LLMs) can derive from text data to offer insights from large collections of documents in a highly interpretable format. This approach aims to bridge the gap between natural language processing (NLP) techniques and economic decision-making, offering a richer, more context-aware understanding of complex economic phenomena.

Continue reading →

Tag: Large Language Models

Narratives under the microscope: what MPC transcripts can tell us about policy thinking

All shocks are different: insights from sentiment and topic analysis using LLMs