Zusammenfassung:
Do financial crises radicalize voters? We study Germany's banking crisis of 1931, when two major banks collapsed and voting for radical parties soared. We collect new data on bank branches and rm-bank connections of over 5,500 firms and show that incomes plummeted in cities affected by the bank failures; connected firms curtailed their payrolls. We further establish that Nazi votes surged in locations exposed to failing Danatbank, led by a prominent Jewish manager and targeted by anti-Semitic Nazi propaganda. Our results suggest a synergy between cultural and economic factors: Danatbank's collapse boosted Nazi support especially in cities with deep-seated anti-Semitism; and the Nazis gained few additional votes in cities exposed to collapsing Dresdner Bank, which was not the target of Nazi hate speech. Danat-exposed and non-exposed cities were similar in their pre-crisis characteristics and exhibited no differential pre-trends; firms borrowing from Danat had lower leverage before the crisis than other rms. Unobservables are unlikely to account for the results.