Words That Unite The World: A Unified Framework for Deciphering Central Bank Communications Globally

TL;DR

This paper introduces the WCB dataset, a comprehensive collection of central bank communications, and benchmarks multiple language models on tasks like stance detection, revealing that aggregated data improves model performance and supporting economic analysis.

Contribution

The paper presents the creation of the largest central bank communication dataset and provides a benchmarking framework for NLP tasks in monetary policy analysis.

Findings

Models trained on aggregated data outperform those trained on individual banks.

Benchmarking results show the effectiveness of various PLMs and LLMs on policy communication tasks.

Human evaluations confirm the economic utility of the proposed framework.

Abstract

Central banks around the world play a crucial role in maintaining economic stability. Deciphering policy implications in their communications is essential, especially as misinterpretations can disproportionately impact vulnerable populations. To address this, we introduce the World Central Banks (WCB) dataset, the most comprehensive monetary policy corpus to date, comprising over 380k sentences from 25 central banks across diverse geographic regions, spanning 28 years of historical data. After uniformly sampling 1k sentences per bank (25k total) across all available years, we annotate and review each sentence using dual annotators, disagreement resolutions, and secondary expert reviews. We define three tasks: Stance Detection, Temporal Classification, and Uncertainty Estimation, with each sentence annotated for all three. We benchmark seven Pretrained Language Models (PLMs) and nine Large Language Models (LLMs) (Zero-Shot, Few-Shot, and with annotation guide) on these tasks, running 15,075 benchmarking experiments. We find that a model trained on aggregated data across banks significantly surpasses a model trained on an individual bank's data, confirming the principle "the whole is greater than the sum of its parts." Additionally, rigorous human evaluations, error analyses, and predictive tasks validate our framework's economic utility. Our artifacts are accessible through the HuggingFace and GitHub under the CC-BY-NC-SA 4.0 license.