TL;DR
Get smart everyday buys 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 more frequent speeches, shorter and more accessible language, and a stronger focus on inflation in the flexible-exchange-rate era.
Swiss National Bank researchers Severin Joho and Enzo Rossi have published a working paper analyzing 1,063 speeches by senior SNB officials from 1928 through 2024. The study finds that speeches became more frequent, shorter and more accessible over time, while inflation took a more prominent place in policy discussion during the era of flexible exchange rates.
The authors say the dataset covers speeches delivered by high-ranking Swiss National Bank officials across nearly a century. Their analysis tracks how often speeches were given, their length and readability, and the themes they addressed. The researchers describe the collection as a novel, comprehensive dataset, though the working-paper summary does not provide a speech-by-speech breakdown in the supplied material.
The paper identifies an upward trend in annual speech frequency beginning in the mid-1950s, with a marked acceleration after Switzerland moved to flexible exchange rates. It also reports a shift toward shorter speeches and increasingly accessible language from the 1960s onward. In the thematic analysis, inflation moved to the forefront in the flexible-exchange-rate era, while cyclical economic development and unemployment received less emphasis.
To compare methods, Joho and Rossi applied traditional Python-based natural-language processing alongside ChatGPT, a generative AI system. The authors say their methodological discussion emphasizes using probabilistic language models alongside deterministic Python frameworks, while accounting for algorithmic pitfalls and unresolved research questions. The supplied summary does not give numerical effect sizes or detailed performance scores for either method.
How SNB Communication Has Shifted
Central-bank speeches are one way officials explain policy and economic conditions to the public and financial markets. A documented shift toward shorter, more accessible language may help readers follow those messages, while the changing frequency of speeches offers a measure of how the institution’s public communication developed. The paper describes patterns over time; it does not establish why each change occurred or measure its effect on policy decisions or audience understanding.
The reported change in subject emphasis also provides a historical view of the SNB’s public policy narrative. The paper links inflation’s increased prominence to the flexible-exchange-rate era, alongside less discussion of cyclical developments and unemployment. That pattern may interest researchers studying central-bank priorities, but speech themes should not be treated as a complete account of the bank’s policy or internal deliberations.
The comparison of text-analysis approaches is relevant beyond this dataset. Researchers increasingly use generative AI to classify and interpret large collections of documents. Joho and Rossi’s approach highlights the value of comparing such systems with rule-based, reproducible tools, rather than assuming that an AI-generated reading is accurate on its own.
central banking history book Swiss National Bank monetary policy
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
The Study’s Century-Long Speech Record
The working paper covers speeches delivered between 1928 and 2024, a period spanning major changes in monetary policy and public communication. The authors specifically report that speech frequency increased from the mid-1950s and accelerated after the transition to flexible exchange rates. The supplied source does not give a precise date for that transition or a detailed account of how the researchers divided the historical periods.
The study examines both measurable features, such as speech length and annual frequency, and language features, including readability and thematic content. Its two analytical approaches serve different roles: Python-based text-mining methods follow defined procedures, while ChatGPT uses a probabilistic language model to process text. The authors present them as complementary, and flag methodological limitations rather than describing either as a definitive substitute for the other.
The source identifies this research as an SNB working paper. Its findings are the authors’ analysis of historical speeches, not necessarily an official policy statement by the central bank. The supplied report summary does not include peer-review status, detailed coding rules or the full results tables.
“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, in the working-paper summary
natural language processing with Python book text analysis
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Details Still Missing From the Summary
The supplied working-paper summary does not report exact year-by-year counts, average speech lengths, readability scores or the size of the thematic shifts. It also does not specify how individual topics were classified, how ChatGPT was prompted, or how closely its outputs matched the Python-based results. Those details are needed to assess the scale and reproducibility of the reported patterns.
The summary says speech frequency accelerated after the transition to flexible exchange rates, but does not establish that the monetary-policy change caused the increase. Nor does it explain why officials’ language became shorter or more accessible. The findings describe patterns in speeches, not changes in the SNB’s underlying decisions, and their relationship to public comprehension remains unreported in the supplied material.
Python data science book pandas NLP text mining
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What to Check in the Full Paper
The next step for readers seeking more detail is to consult the full SNB working paper, which may provide the dataset description, analytical procedures, comparisons between Python and ChatGPT, and supporting results. The source summary says the authors also outline open questions for future research, but does not list a timetable or specific planned follow-up studies.
Further work could test whether the reported trends hold under different topic classifications or language measures, and clarify how model choices affect results. For now, the reported conclusions cover speeches through 2024; the source gives no information about later speeches or a planned update to the dataset.
large language models for research and text analysis 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 from 1928 to 2024, examining their frequency, length, readability and themes.
What changes in SNB speeches did the authors report?
The authors report that speeches became more frequent, especially after the transition to flexible exchange rates, and that they grew shorter and more accessible. They also say inflation became a more prominent theme.
How did the researchers use ChatGPT?
They benchmarked ChatGPT against Python-based text-mining methods as part of their analysis. The supplied summary does not provide detailed accuracy scores or explain the exact prompts used.
Does the study show why SNB communication changed?
No. The summary reports historical patterns but does not establish why speech frequency, length, accessibility or subject emphasis changed, or whether those changes affected policy or public understanding.
Is the working paper an official SNB policy announcement?
No. It is a research paper analyzing historical speeches, not a new monetary-policy decision or policy announcement. Its conclusions should be read as findings from the authors’ study.
Source: primary
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
