A bank with weak financial health has a nearly 100 basis point probability of experiencing a run in a given year, compared to just 25 basis points for strong banks, according to new research published July 7, 2026 by the Federal Reserve Bank of New York. The study, authored by Sergio Correia of the Federal Reserve Bank of Richmond, Stephan Luck of the New York Fed, and Emil Verner of MIT Sloan School of Management, analyzed more than 3,000 bank runs from 1863 to 1934 to settle a longstanding debate: are runs the key trigger that turns small shocks into major crises, or are they mainly symptoms of deeper problems in the financial system? The report concludes that poor bank fundamentals are central to explaining both when runs occur and when they cause severe economic damage.
The researchers found that a bank has a 38 percent probability of failing if it experiences a run, though the annual unconditional probability of bank failure is just 0.85 percent. This means runs result in more survivals than failures overall. Banks in the lowest decile of financial health—measured by capitalization, profitability, and funding structure—have a 63 percent probability of failing when subject to a run. In stark contrast, the strongest banks in the top decile essentially never fail when runs occur. The study also revealed that runs on strong banks don't significantly harm local manufacturing activity within eighteen months, while runs with bank failures predict manufacturing declines exceeding 5 percent. The database covers the National Banking Era panics and the Great Depression, analyzing newspaper records combined with bank balance sheets, macroeconomic data, and local business failures.
The report finds that weak bank fundamentals are necessary for runs to result in bank failures and economic disruption. "Our findings lend little support to the view that small shocks by themselves can result in widespread banking panics that cause major economic downturns," the authors write. Instead, the evidence indicates that poor fundamentals are necessary for runs to generate severe local financial disruptions. The researchers argue that runs should be seen as a trigger for bank failures and crises, but insolvent banks are necessary for this trigger to devastate the banking system and the economy. While runs can hit both weak and strong banks, only those with poor financial health translate withdrawals into widespread damage.
The report explains that strong banks survive runs through a series of defensive measures documented in contemporary newspapers. Banks accommodate withdrawals and signal strength—sometimes with dramatic gestures like delivering truckloads of cash. Owners inject additional equity or borrow from other banks to obtain liquidity. Many banks invoke 30- or 60-day notice rules for savings deposits to slow withdrawals, while some fully suspend convertibility. About 10 percent of runs involved clearinghouses providing loans or examinations to determine solvency. These measures severed the link between illiquidity and insolvency for healthy institutions. The historical setting matters because runs were far more common before government interventions like deposit insurance and lending of last resort, making the data particularly valuable for understanding crisis dynamics. The authors' conclusion tempers fears that minor shocks can spiral into catastrophic outcomes through self-fulfilling panic, pointing instead to underlying bank weakness as the critical factor determining whether financial turbulence becomes economic disaster.