Bank Runs With and Without Bank Failure

TL;DR

This study analyzes historical bank runs in the U.S. from 1863 to 1934 using large language models to identify and examine their causes, effects, and the role of bank fundamentals in crisis outcomes.

Contribution

It introduces a novel dataset of bank runs identified via language models and provides micro-level insights into the causes and consequences of bank runs in historical context.

Findings

Runs are more common in weak banks but also occur in strong banks.

Bank failures are mainly in banks with poor fundamentals.

Runs on weak banks lead to larger economic declines.

Abstract

We study the causes and consequences of bank runs using a novel dataset of bank runs in the United States from 1863 to 1934. Applying large language models to historical newspapers, we identify 3,421 runs on individual banks. The resulting series aligns closely with narrative chronologies of U.S. banking crises and provides rich new micro-level information on runs. Runs are considerably more likely in weak banks but also occur in strong banks, especially in response to negative news about the real economy or the broader banking system. However, runs typically result in failure only for banks with poor fundamentals. Strong banks survive runs through various mechanisms, including interbank cooperation and suspension of convertibility. At the local level, runs on banks with poor fundamentals translate into substantially larger declines in deposits, lending, and manufacturing activity than runs on strong banks. Our findings imply that poor fundamentals are central to explaining both when runs occur and when they have severe economic effects, tempering the view that small shocks can generate discontinuous jumps to bad equilibria through self-fulfilling run dynamics.