Title: Analysis of central bank communication using advanced text modeling methods

Funder:  Austrian National Bank (OeNB), (link) 

Project Team: Paul Hofmarcher (University Salzburg, Principal Investigator), Bettina Grün (WU Vienna, Co-Investigator), Niko Hauzenberger (University Salzburg, Co-Investigator) 

Project Funded Staff: Niko Hauzenberger (July 2022-January 2023),  Jan Vavra (January 2023-2024), Jolanda Heinemeier (2024-2026)

Appropriate communication strategies from central banks are considered to be important tools for the effective implementation of monetary policy. Especially since the financial crisis, where many central banks were tied to the zero lower bound, good communication is essential to compensate for this limitation.

We analyze central bank speeches spanning more than five decades to examine different communication strategies and their effects on macroeconomic indicators, taking into account differences in time, region and actors. The project includes the three main contributions. First, building on recently developed text analysis models, we develop extensions and adaptations so that they are particularly suitable for central bank speeches. Second, we apply the existing and newly developed models to a new data set that contains more than 16,000 speeches by central bankers, which are available on the BIS website. Third, we use covariates derived from the text analysis to fit vector autoregressions to these covariates as well as macroeconomic indicators and thus carry out a structural analysis of a communication shock and to check whether this improves the predictive quality for macroeconomic indicators.

Final Symposium: The final symposium of this workshop will take place at May, 28th 2026 @ WU Vienna 

Publications: 

• Jan Čapek, Jesús Crespo Cuaresma,  Niko Hauzenberger,  Vlastimil Reichel; Macroeconomic forecasting in the euro area using predictive combinations of DSGE models, International Journal of Forecasting (2022 forthcoming) link 

• Bernd Prostmaier, Bettina Grün and Paul Hofmarcher. Variational Seeded Topic Model: Leveraging domain knowledge to fit semi-supervised topic models in large scale data environments, Knowledge Based Systems, 327; 2025 paper 

• Jan Vavra, Bettina Grün and Paul Hofmarcher. Evolving Voices Based on Temporal Poisson Factorisation,  accepted  Statistical Modelling 2025, arXiv 

•  Paul Hofmarcher,  Jan Vavra,  Sourav Adhikari , Bettina Grün. Revisiting Group Differences in High-Dimensional Choices: Method and Application to Congressional Speech,  Journal of Applied Econometrics, 2025, replication code 

• Pierre Siklos, Martin Feldkircher, One Money, One Voice? Evaluating Ideological Positions of Euro Area Central Banks. European Journal of Political Economy, 85; 2024 paper 

Working Papers:

• Niko Hauzenberger, Florian Huber, Karin Klieber and Massimiliano Marcellino; Enhanced Bayesian Neural Networks for Macroeconomics and Finance, link 

• Niko Hauzenberger, Florian Huber, Massimiliano Marcellino, and NicoPetz; Gaussian Process Vector Autoregressions and Macroeconomic Uncertainty, link 

• Martin Feldkircher, Paul Hofmarcher and Pierre L. Siklos; One Money, One Voice? Evaluating Ideological Positions of Euro Area Central Banks

• A Structural Text-Based Scaling Model,  (with Jan Vavra, Bernd Prostmaier, Bettina Grün, R&R Political Analysis)

• ECB Communication Complexity, Sentiment, and Inflation Expectations: What Matters Most? (with Pierre Siklos, Martin Feldkircher, Benrd Prostmaier; submitted)

• Poisson Factorization Based Topic Models (with Bernd Prostmaier,  Karina  Pekarek-Kostka, Bettina Grün)

• Covariate-Informed Seeded Poisson Factorization: Combining Seed-Guided Topic Discovery and Covariate-Aware Topic Prevalence (with Bernd Prostmaier, Matthias Assenmacher,  Bettina Grün)