TL;DR
Get your next haul delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
A Swiss National Bank working paper by Severin Joho and Enzo Rossi examines 1,063 speeches by senior SNB officials from 1928 to 2024. It reports that speeches became more frequent, shorter and more accessible, while inflation gained prominence in policy discussion during the flexible exchange-rate era.
The Swiss National Bank (SNB) has published a working paper analyzing how its senior officials’ speeches changed over nearly a century, reporting a rise in speech frequency, a shift toward shorter and more accessible language, and greater attention to inflation in the flexible exchange-rate era. Authors Severin Joho and Enzo Rossi assembled 1,063 speeches from 1928 to 2024 and compared conventional Python-based text analysis with ChatGPT-assisted analysis.
The authors say the frequency of annual speeches followed an upward trajectory from the mid-1950s, with a more pronounced acceleration after Switzerland moved to flexible exchange rates. The paper also finds a broad change in style: speeches became shorter and their language increasingly accessible from the 1960s onward. The source summary does not provide specific yearly counts, readability scores or average speech lengths.
The analysis also identifies a change in subject matter. Inflation moved to the forefront of the SNB’s thematic discourse in the flexible exchange-rate era, while cyclical economic development and unemployment received less emphasis. These are findings about the content of speeches, not measures of the bank’s actions or the effects of its policies.
Alongside the historical analysis, Joho and Rossi compare traditional Python-based natural-language processing with generative AI using ChatGPT. They describe the methods as complementary and discuss algorithmic pitfalls and open questions for future research. The summary does not state that ChatGPT replaces conventional text-mining methods.
How SNB Communication Changed
The study offers a long-run view of how a central bank communicates with the public and how its public-facing priorities shift alongside changes in monetary policy conditions. The reported move toward shorter, more accessible speeches may help explain changes in the form of official communication, while the increased prominence of inflation marks a change in the topics emphasized by SNB officials.
For researchers and readers following central-bank messaging, the paper also speaks to a practical question: how to analyze large historical collections of text without treating an AI-generated interpretation as definitive. Its method comparison highlights the value of using language models alongside reproducible, rule-based tools, while remaining attentive to the limits of each approach.
natural language processing with Python book text mining
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
A Century of SNB Speeches
The study covers speeches delivered between 1928 and 2024 by high-ranking SNB officials. That span allows the authors to compare communication across different monetary-policy settings, including the period before and after the transition to flexible exchange rates. The supplied report summary does not give a detailed chronology of that transition or name individual speeches used in the analysis.
The project combines quantitative tracking with examination of speech length, readability and themes. Its dataset contains 1,063 speeches, which the authors describe as a novel and comprehensive collection. The working paper’s stated focus is both the evolution of SNB rhetoric and the methods used to study it, rather than an assessment of current policy decisions.
“The frequency of annual speeches has followed an upward trajectory since the mid-1950s, accelerating notably after the transition to flexible exchange rates.”
— Severin Joho and Enzo Rossi, authors of the SNB working paper
Python text analysis and NLP reference book
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Limits of the Published Summary
The supplied summary does not include the underlying annual speech counts, quantitative readability measures, or the detailed coding used to compare themes. It therefore does not establish the size of each change or how results vary across individual decades. It also does not specify how ChatGPT was prompted, which model version was used, or how AI-generated classifications were checked against the Python-based results.
The paper’s findings describe patterns in public speeches. They do not, on their own, show why officials changed their style or topic emphasis, or establish a direct relationship between speech patterns and public understanding or policy outcomes. The source identifies methodological pitfalls and open research questions but does not detail all of them in the provided summary.
large language models for social science research book
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Further Testing of Text Analysis
The authors point to open questions for future research on using large language models to analyze central-bank communication. The working paper’s publication makes its dataset and methodological choices relevant points for further scrutiny, though the supplied material does not announce a follow-up study, release schedule or next research milestone.
Readers seeking to evaluate the findings in detail will need the full working paper, including its methods and results tables. Until those details are considered, the broad trends reported by the authors can be described, but their precise scale and the relative performance of the two analytical approaches remain unclear from the summary alone.
computational text analysis research methods book
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What does the SNB working paper study?
It analyzes 1,063 speeches by high-ranking Swiss National Bank officials delivered between 1928 and 2024, examining frequency, length, readability and themes.
What changes in speech style did the authors report?
The authors report that speeches became shorter and increasingly accessible from the 1960s onward. The supplied summary does not give precise length or readability figures.
How did the topics discussed in speeches change?
The paper says inflation became more prominent in the flexible exchange-rate era, while cyclical economic development and unemployment received less emphasis.
Did the study find that ChatGPT is better than Python-based analysis?
The summary does not claim that ChatGPT is superior. The authors describe generative AI and traditional Python-based text analysis as complementary and discuss limitations and methodological questions.
Does the paper explain why SNB speeches changed?
Not in the supplied summary. It reports historical patterns, including an acceleration in speech frequency after the move to flexible exchange rates, but does not establish why the changes occurred.
Source: primary
Fall Picks
fall essentials
As an affiliate, we earn on qualifying purchases.
