Publications
Deakin, S., & Shuku, L. (2025). Exploring computational approaches to law: the evolution of judicial language in the Anglo‐Welsh poor law, 1691–1834. Journal of Law and Society. https://doi.org/10.1111/jols.12521
Abstract
The use of natural language processing (NLP) and machine learning (ML) to analyse the structure of legal texts is a fast-growing field. While much attention has been devoted to the use of these techniques to predict case outcomes, they have the potential to contribute more broadly to research into the nature of legal reasoning and its relationship to social and economic change. In this paper, we use recently developed NLP and ML methods to test the claim that judicial language is systematically shaped by economic shocks deriving from the business cycle and by long-run trends in the economy associated with technological change and industrial transition. Focusing on cases decided under the Anglo-Welsh poor law between the 1690s and 1830s, we show that the terminology used to describe the right to poor relief shifted over time according to economic conditions. We explore the implications of our results for the poor law, the theory of legal evolution, and socio-legal research methods.
Awarded Gavin C Reid Prize for the Best Paper by a Centre for Business Research (CBR) Judge Business School Early Career Researcher 2025
Chortareas, G., Papailias, F., & Shuku, L. (2025). Does Central Bank Talk Matter for Forecasting? Evidence From Speeches of the BoE, ECB, and Fed. International Journal of Finance & Economics. https://doi.org/10.1002/ijfe.3136
Supplementary Information can be found here, Online Appendix
Abstract
This paper explores the information content of the untargeted narratives of the Bank of England (BoE), the European Central Bank (ECB), and the Federal Reserve (Fed) and whether it has the potential to improve forecasting performance. We apply the Latent Dirichlet Allocation (LDA) method to extract topics from the corpus of text data. We then evaluate the impact of these central bank officials' speeches on macroeconomic and financial variables forecasting from 1997 to 2018. Our results suggest that the forecasting model, incorporating information from speeches, produces estimates with a lower forecasting error for several variables in the UK, the EU, and the US. For certain variables, the forecast improvement is more pronounced after the global financial crisis of 2008.
Data collection: English Poor Law Cases, 1690-1815
(with S. Deakin and V. Cheok) (2024)
[Available here: link]
Description
This dataset of historical poor law cases was created as part of a project aiming to assess the implications of the introduction of Artificial Intelligence (AI) into legal systems in Japan and the United Kingdom. The project was jointly funded by the UK’s Economic and Social Research Council, part of UKRI, and the Japanese Society and Technology Agency (JST), and involved collaboration between Cambridge University (the Centre for Business Research, Department of Computer Science and Faculty of Law) and Hitotsubashi University, Tokyo (the Graduate Schools of Law and Business Administration). As part of the project, a dataset of historic poor law cases was created to facilitate the analysis of legal texts using natural language processing methods. The dataset contains judgments of cases which have been annotated to facilitate computational analysis. Specifically, they make it possible to see how legal terms have evolved over time in the area of disputes over the law governing settlement by hiring.
Working Papers
Understanding Communication Strategies of the Three Major Central Banks Through Speeches
This paper provides a descriptive analysis of the speeches of central banks with a broad and non-expert audience. It provides a comprehensive understanding of the sort of information that central bankers communicate through speeches. More specifically, the in-depth analyses of the textual data offer a comparative analysis of the speech content between three major central banks, the BoE, the Fed, and the ECB, in addition to the within-bank analysis at the aggregate and head-of-the-bank level for the period between 1998 to 2018. Methodologically, this paper is exclusively focused on textual data analysis that involves Natural Language Processing (NLP) techniques, specifically the word frequency methods and sentiment analysis through a dictionary approach. The findings suggest considerable differences in communication patterns through speeches between the three banks in terms of the diversity of themes discussed, the information disclosure, and, overall, the volume of talks pertaining to specific events, such as the global financial crisis of 2008. This indicates that talks are increasingly used to openly discuss economic activity with the general public of three banks. Still, the intensity of using such a communication channel and the reaction expressed vis-à-vis the change in the economic state is different across banks.
The Impact of Central Bank Communication Sentiment on Professional Forecasters’ Predictions During the Periods of Forward Guidance
This paper evaluates the impact of sentiment change during forward guidance periods on the forecast accuracy and convergence of the main macroeconomic measures for the ECB. It specifically assesses how the ECB’s different types of forward guidance used since 2013 impact inflation, growth, and unemployment estimates. Additionally, it expands on evaluating how the sentiment change in the talks by the ECB from 1998 to 2021, measured through six different dictionaries, impacts forecast accuracy and convergence for all three variables. The study uses the quarterly forecast data from the Survey of Professional Forecasters. The results show that growth and unemployment estimates become more accurate during time-contingent guidance, whereas forecast convergence is mainly noted during data-driven guidance/state-contingent guidance. Changes in the tone of the speeches explicitly improve the accuracy of growth estimates and yield a lower standard deviation of point forecast estimates for growth and unemployment forecasts. Results mainly hold for the change in sentiment measured through the dictionaries built from economic text with manual classification. Results with the interaction term of sentiment and forward guidance dummy yield better results on forecast accuracy and convergence, meaning that a change in the tone of the talks during forward guidance is more informative for predictions of macroeconomic measures as compared to the period without official guidance. The study does not find significant results for inflation estimates, especially with regard to the evaluation of forecast accuracy.
Other
Journal of Law and Society blog post - Meet the JLS Authors: Using computational techniques to explore the Anglo-Welsh poor law
[Available here: link]