# Correia, Luck, and Verner explain in Liberty Street Economics how AI helps map bank runs [Canonical reading](https://escapemeridian.com/readings/correia-luck-and-verner-explain-in-liberty-street-economics-how-4da87f61) On July 7, 2026, Sergio Correia, Stephan Luck, and Emil Verner describe using large language models to identify bank runs in historical newspaper articles, where official records often show bank closures but not whether a run preceded them. Their approach searches for candidate articles, then uses models to extract events and distinguish runs, suspensions, and failures. The authors say this helps capture varied descriptions that keyword matching can miss; they compare outputs from multiple models and report high agreement with hand-coded samples. The resulting database contains more than 3,000 run episodes from 1863 through 1934. The authors’ central distinction is that most recorded runs did not involve failure, even though most distress episodes in the dataset involved a bank failure. They also describe runs spreading during major crises and local distress appearing beyond the national crisis narrative. The database makes newspaper accounts available for readers to examine, while giving researchers material for questions that were difficult to study with regulatory records alone. The authors say their companion post uses the data to investigate when and why runs led to failure and the implications for policy discussions. The post does not specify the size of the hand-coded samples used for validation. Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims. ## Original source Federal Reserve Bank of New York · Article Original source date: Jul 7, 2026 (2026-07-07T14:00:00.000Z). Date basis: Date supplied by the publisher’s feed. Added to Escape Meridian: 2026-11-07T04:33:24.537Z. First observed: 2026-10-30T02:30:03.519Z. Retrieved: 2026-10-30T02:31:44.621Z. - [Using AI to Let History Speak About Bank Runs](https://libertystreeteconomics.newyorkfed.org/2026/07/using-ai-to-let-history-speak-about-bank-runs/) ## Limits Text as retrieved on October 30, 2026; originally published July 7, 2026. The post reports high agreement with hand-coded samples but does not state their size. The database draws on digitized historical newspaper articles; the post does not establish how many bank-run episodes went unreported in those articles. ## Open question How well do the newspaper-based classifications capture bank runs that were not reported in the digitized articles? ## Source-stated Correia, Luck, and Verner describe using large language models to extract bank-run information from historical newspaper articles and build a database. > In a new paper, we use large language models (LLMs) to extract information on bank runs from millions of digitized historical newspaper pages - [Supporting source](https://libertystreeteconomics.newyorkfed.org/2026/07/using-ai-to-let-history-speak-about-bank-runs/) ## Source-stated The authors say their approach can identify varied descriptions of runs that rigid keyword searches might miss. > The language model can understand these descriptions in context, something that rigid keyword-based approaches would miss. - [Supporting source](https://libertystreeteconomics.newyorkfed.org/2026/07/using-ai-to-let-history-speak-about-bank-runs/) ## Source-stated The authors report high agreement between model classifications and hand-coded samples, but do not give the samples’ size in this post. > We further validate our LLM classifications against hand-coded samples and find high agreement - [Supporting source](https://libertystreeteconomics.newyorkfed.org/2026/07/using-ai-to-let-history-speak-about-bank-runs/) ## Source-stated The database contains more than 3,000 bank-run episodes during the stated sample period. > During the 1863-1934 period, our dataset contains more than 3,000 bank run episodes. - [Supporting source](https://libertystreeteconomics.newyorkfed.org/2026/07/using-ai-to-let-history-speak-about-bank-runs/) ## Source-stated The authors find more runs without failure than runs with failure, while most newspaper-recorded distress episodes involve a bank failure. > We find 1,906 runs without failure and 1,515 runs with failure. We also find 13,069 episodes involving a suspension, and 10,330 episodes involving a failure. Thus, the majority of distress episodes recorded in newspapers involve a bank failure. - [Supporting source](https://libertystreeteconomics.newyorkfed.org/2026/07/using-ai-to-let-history-speak-about-bank-runs/) ## Source-stated The authors say their companion post uses the database to investigate when and why runs led to bank failure and discuss policy implications. > In our companion post, we show how we use the data to investigate when and why runs have historically led to bank failure, and discuss how these insights affect policy discussions. - [Supporting source](https://libertystreeteconomics.newyorkfed.org/2026/07/using-ai-to-let-history-speak-about-bank-runs/) ## Source and contributor context Federal Reserve Bank of New York · Publisher The publisher is identified here. No individual author has been inferred. ## Source media Media retrieval status: unchecked. ## Terms and references Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement. ## Large language model A language model with many learned numerical settings, used to process and generate language. - [Definition reference](https://developers.google.com/machine-learning/glossary#large-language-model) ## Editorial provenance Version 1. Publication ID: 5dd3c258-e36e-7834-6083-63c1aca07806. Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5. Source check: passed by anthropic/claude-opus-5.5 on 2026-11-07T04:30:40.494Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: bridge\_admitted\_by\_model. Observation: b1dd3595e4378865412bcd2d9183213d1f332d1784644db693bc5f99ad2bf5ac. Source text hash: 00f0960a5e77546d66d42cf07978af690b9f158e4b6dc7402b9411caa64d9d7a. ## Revision and source-check history ### Version 1 — 2026-11-07T04:33:24.537Z On July 7, 2026, Sergio Correia, Stephan Luck, and Emil Verner describe using large language models to identify bank runs in historical newspaper articles, where official records often show bank closures but not whether a run preceded them. Their approach searches for candidate articles, then uses models to extract events and distinguish runs, suspensions, and failures. The authors say this helps capture varied descriptions that keyword matching can miss; they compare outputs from multiple models and report high agreement with hand-coded samples. The resulting database contains more than 3,000 run episodes from 1863 through 1934. The authors’ central distinction is that most recorded runs did not involve failure, even though most distress episodes in the dataset involved a bank failure. They also describe runs spreading during major crises and local distress appearing beyond the national crisis narrative. The database makes newspaper accounts available for readers to examine, while giving researchers material for questions that were difficult to study with regulatory records alone. The authors say their companion post uses the data to investigate when and why runs led to failure and the implications for policy discussions. The post does not specify the size of the hand-coded samples used for validation. ## Original source Federal Reserve Bank of New York · Article Original source date: Jul 7, 2026 (2026-07-07T14:00:00.000Z). Date basis: Date supplied by the publisher’s feed. Added to Escape Meridian: 2026-11-07T04:33:24.537Z. First observed: 2026-10-30T02:30:03.519Z. Retrieved: 2026-10-30T02:31:44.621Z. - [Using AI to Let History Speak About Bank Runs](https://libertystreeteconomics.newyorkfed.org/2026/07/using-ai-to-let-history-speak-about-bank-runs/) ## Limits Text as retrieved on October 30, 2026; originally published July 7, 2026. The post reports high agreement with hand-coded samples but does not state their size. The database draws on digitized historical newspaper articles; the post does not establish how many bank-run episodes went unreported in those articles. ## Open question How well do the newspaper-based classifications capture bank runs that were not reported in the digitized articles? ## Source-stated Correia, Luck, and Verner describe using large language models to extract bank-run information from historical newspaper articles and build a database. > In a new paper, we use large language models (LLMs) to extract information on bank runs from millions of digitized historical newspaper pages - [Supporting source](https://libertystreeteconomics.newyorkfed.org/2026/07/using-ai-to-let-history-speak-about-bank-runs/) ## Source-stated The authors say their approach can identify varied descriptions of runs that rigid keyword searches might miss. > The language model can understand these descriptions in context, something that rigid keyword-based approaches would miss. - [Supporting source](https://libertystreeteconomics.newyorkfed.org/2026/07/using-ai-to-let-history-speak-about-bank-runs/) ## Source-stated The authors report high agreement between model classifications and hand-coded samples, but do not give the samples’ size in this post. > We further validate our LLM classifications against hand-coded samples and find high agreement - [Supporting source](https://libertystreeteconomics.newyorkfed.org/2026/07/using-ai-to-let-history-speak-about-bank-runs/) ## Source-stated The database contains more than 3,000 bank-run episodes during the stated sample period. > During the 1863-1934 period, our dataset contains more than 3,000 bank run episodes. - [Supporting source](https://libertystreeteconomics.newyorkfed.org/2026/07/using-ai-to-let-history-speak-about-bank-runs/) ## Source-stated The authors find more runs without failure than runs with failure, while most newspaper-recorded distress episodes involve a bank failure. > We find 1,906 runs without failure and 1,515 runs with failure. We also find 13,069 episodes involving a suspension, and 10,330 episodes involving a failure. Thus, the majority of distress episodes recorded in newspapers involve a bank failure. - [Supporting source](https://libertystreeteconomics.newyorkfed.org/2026/07/using-ai-to-let-history-speak-about-bank-runs/) ## Source-stated The authors say their companion post uses the database to investigate when and why runs led to bank failure and discuss policy implications. > In our companion post, we show how we use the data to investigate when and why runs have historically led to bank failure, and discuss how these insights affect policy discussions. - [Supporting source](https://libertystreeteconomics.newyorkfed.org/2026/07/using-ai-to-let-history-speak-about-bank-runs/) ## Source and contributor context Federal Reserve Bank of New York · Publisher The publisher is identified here. No individual author has been inferred. ## Source media Media retrieval status: unchecked. ## Terms and references Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement. ## Large language model A language model with many learned numerical settings, used to process and generate language. - [Definition reference](https://developers.google.com/machine-learning/glossary#large-language-model) ## Editorial provenance Version 1. Publication ID: 5dd3c258-e36e-7834-6083-63c1aca07806. Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5. Source check: passed by anthropic/claude-opus-5.5 on 2026-11-07T04:30:40.494Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: bridge\_admitted\_by\_model. Observation: b1dd3595e4378865412bcd2d9183213d1f332d1784644db693bc5f99ad2bf5ac. Source text hash: 00f0960a5e77546d66d42cf07978af690b9f158e4b6dc7402b9411caa64d9d7a. ### passed — 2026-11-07T04:30:40.494Z — anthropic/claude-opus-5.5 Publication: 5dd3c258-e36e-7834-6083-63c1aca07806 attribution — pass: The reading correctly credits Sergio Correia, Stephan Luck and Emil Verner, the post's bylined authors, and Liberty Street Economics, the Federal Reserve Bank of New York blog. The methods and findings are framed as the authors' own descriptions and results. This is a single authored post, not a listing. > Sergio Correia, Stephan Luck, and Emil Verner summary — pass: The title, summary and question faithfully convey the post. The authors used LLMs to mine newspapers for bank runs that official records miss. The reading explains the keyword-then-model pipeline, the validation, the more-than-3,000 episodes from 1863-1934, and the key run-versus-failure distinction. It also covers geographic and crisis patterns and the companion post. LLM is defined beside the reading, and the context about regulatory records explains why the work matters. The question follows from newspaper-coverage limits without leading. > But such records do not indicate whether a run occurred before failure. date — pass: The July 7, 2026 date matches the post's dateline, the citation and the UTC day of originalPublished. The page was modified on 2026-08-07, but every cited passage is plainly part of the body of the original article. The reading does not use the later navigation or related-story links and contains no hindsight framing. Kind is publisher-page, so no archive disclosure is needed. > July 7, 2026 support — pass: Each claim is supported verbatim: the LLM extraction, the advantage over keyword matching, the validation against hand-coded samples, the episode counts, the run-versus-failure finding and the companion post. The noted gap is accurate, since the post gives no hand-coded sample size. The stated uses (public browsing, research questions) are supported. Body text appears original despite the later edit flag. > We find 1,906 runs without failure and 1,515 runs with failure. media — pass: No media was selected and no verified media candidates were supplied, so nothing is misidentified or irrelevant. ## Recorded references Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter. No public references recorded here yet. [Reference data (JSON)](/readings/4da87f61c464b09a9bff3d095d1e1c4b3ec662cab1b1e93b3321cc9daf6a6990/references.json) ## Corrections and challenges [Send a correction or an opposing source](https://escapemeridian.com/corrections/) [JSON record](https://escapemeridian.com/readings/correia-luck-and-verner-explain-in-liberty-street-economics-how-4da87f61.json)