How do violent conflicts shape cross-border lending? Using data on syndicated loans by 14,021 creditors to firms in 179 countries (1989-2020), we document a dual effect: foreign banks reduce overall lending relative to domestic banks but significantly increase financing to military and dual-use sectors during conflicts. This reallocation is stronger among lenders less specialized in the conflict country, more specialized in military lending, and domiciled in politically non-aligned nations. Effects are geographically contained and temporally limited, dissipating post-conflict. Our findings reveal how global banks strategically redirect credit toward military sectors during armed conflicts, despite reducing overall country exposure.

We study bank runs using a novel historical cross-country dataset that covers 184 countries since 1800 and combines a new narrative chronology with statistical indicators of bank deposit withdrawals. We document the following facts: (i) the unconditional likelihood of a bank run is 1.9%, and that of significant deposit withdrawals is 12.5%; (ii) systemic bank runs---those that are accompanied by deposit withdrawals---are associated with substantially larger output losses than non-systemic runs or deposit contractions alone; (iii) bank runs are contractionary even when they are not triggered by fundamental causes, banks are well-capitalized, and there is no evidence of a crisis or widespread failures in the banking sector; (iv) in both historical and contemporary episodes, depositors tend to run on highly leveraged banks, which leads to a credit crunch and a reallocation of deposits across banks; and (v) liability guarantees are associated with lower output losses after systemic runs, while having a lender of last resort or deposit insurance reduces the probability of a run becoming systemic. Overall, our findings highlight a key role of sudden bank liability disruptions in economic fluctuations, over and above other sources of financial fragility.

This paper introduces a novel text-based measure of banking regulation intensity, the Bank Regulation Index (BRI), constructed from historical newspaper coverage spanning nearly a century of U.S. banking history. While periods of deregulation correspond with immediate improvements in bank performance, these periods are followed by weaker fundamentals and increased instability over longer horizons. The BRI is decomposed into distinct regulatory topics using Latent Dirichlet Allocation (LDA), and the credit-related topic exhibits the strongest association with subsequent banking sector distress. This predictive relationship remains robust when controlled for established indicators such as credit growth and financial liberalization measures. These findings demonstrate the value of text-based approaches in quantifying topic-level regulatory intensity and tracking its relationship with banking sector stability across different time horizons.