Stephan Luck on What History can Teach Us about Financial Stability

Jul 13, 2026
Stephan Luck is a Financial Research Advisor at the Federal Reserve Bank of New York. In Stephan’s first appearance on the show, he discusses how the 2008 Great Financial Crisis shaped his career, how he and his coauthors leverage LLMs to comb through massive amounts of historical data, what this data can teach us about responding to bank failures, what people get wrong when they try and make historical analogies to the GENIUS era, the lessons we can learn from the German hyperinflation, and much more.
Watch the full length video on our new YouTube Channel!
Check out the transcript for this week’s episode, now with links.
Recorded on June 1st, 2026
Subscribe to David's Substack: Macroeconomic Policy Nexus
Follow David on X: @DavidBeckworth
Follow the show on X: @Macro_Musings
Check out our Macro Musings merch!
Timestamps
00:00:00 - Intro
00:02:55 - Stephan’s Career
00:11:21 - Bank Failures
00:28:19 - Policy Implications of Bank Failures
00:37:21 - National Banking System
00:46:30 - German Hyperinflation
00:57:06 - Outro

Facts Only

* Stephan Luck is a Financial Research Advisor at the Federal Reserve Bank of New York.
* The discussion covers the impact of the 2008 Great Financial Crisis on his career.
* The speakers leverage LLMs to comb through historical data.
* The topic includes responses to bank failures, making historical analogies, and German hyperinflation.
* The recording date is June 1st, 2026.

Executive Summary

Stephan Luck, a Financial Research Advisor at the Federal Reserve Bank of New York, discusses the influence of the 2008 Great Financial Crisis on his career. He details how he and coauthors utilize Large Language Models (LLMs) to analyze historical data related to bank failures. The discussion covers what this data reveals about responding to bank failures, misconceptions regarding historical analogies like the GENIUS era, and lessons derived from German hyperinflation. The content was recorded on June 1st, 2026.

Full Take

The discussion frames the utility of advanced data synthesis tools like LLMs in understanding complex economic events, specifically focusing on systemic failures like bank collapses and periods of extreme monetary instability such as hyperinflation. The pattern observed is a move from specific historical case studies (2008 crisis, German inflation) to abstract lessons about systemic resilience. This suggests an underlying assumption that historical patterns are directly translatable into actionable policy wisdom, which warrants scrutiny regarding the limits of analogical reasoning when dealing with evolving institutional structures. The implication for agency is whether relying on LLM-derived analogies risks obscuring novel causal mechanisms specific to current financial environments. Further inquiry should focus on the mechanism by which LLMs facilitate this synthesis and what blind spots arise when applying historical frameworks to contemporary policy challenges.

Sentinel — Human

Confidence

This text exhibits strong characteristics of promotional media designed to advertise a video or podcast episode rather than deep analytical writing; it reads like curated metadata presented by a human entity.

Signals Detected

Sentence length variance shows some natural flow; phrasing is promotional rather than purely informational.

The text functions primarily as an advertisement/summary of a media product rather than deep analytical prose, lacking the emotional or idiosyncratic voice of pure AI synthesis.

Features clear promotional elements (calls to action, links) embedded within what appears to be a transcript context.

The structure strongly suggests a standard podcast/video promotion format, which is easily replicable but contextually plausible for media promotion.

Human Indicators

The inclusion of specific creator names, handles (@DavidBeckworth), and branded elements (Substack, YouTube channel) points toward human production targeting an audience, not raw content generation.