Ornithologist: Towards Trustworthy "Reasoning" about Central Bank Communications

TL;DR

This work tackles measuring hawkishness and dovishness in central bank communications with a transparent, explainable approach. Ornithologist combines a hand-built economic taxonomy with a retriever and a CFG-constrained generator to produce sentence- and paragraph-level classifications accompanied by explicit reasoning traces. Validation shows competitive alignment with policy signals and market expectations, with convergent validity against other Hawk-Dove measures and evidence that the scores relate to future policy stance and market forecasts. The method enhances interpretability, reduces hallucination risk, and offers easy adaptation to other textual sources and contexts, supporting broader use in policy analysis and scenario exploration.

Abstract

I develop Ornithologist, a weakly-supervised textual classification system and measure the hawkishness and dovishness of central bank text. Ornithologist uses ``taxonomy-guided reasoning'', guiding a large language model with human-authored decision trees. This increases the transparency and explainability of the system and makes it accessible to non-experts. It also reduces hallucination risk. Since it requires less supervision than traditional classification systems, it can more easily be applied to other problems or sources of text (e.g. news) without much modification. Ornithologist measurements of hawkishness and dovishness of RBA communication carry information about the future of the cash rate path and of market expectations.

Figures (6)

• Figure 1: Ornithologist system
• Figure 2:
• Figure 3:
• Figure 4:
• Figure 5: Ornithologist reporting