Knowledge that Transforms

This paper identifies a novel channel—corporate fraud—through which local firms protect themselves during anti-mafia periods. We initially explore the impact of the anti-mafia campaign on mafia-like forces captured by accepted mafia-like criminal cases. The results indicate that the average number of accepted mafia-like criminal cases increases by 10.03%. More importantly, the accepted mafia-like criminal cases committed by firms also exhibit a notable increase, mainly within the first 30 months of the anti-mafia campaign. We then theorize that the anti-mafia campaign may have two contradictory effects on corporate fraud: deterrence and self-protection. The main results show that the self-protection effect prevails over the deterrence effect and that firms are more likely to commit fraud to strategically protect themselves from potential losses. The baseline findings reveal that the fraud occurrence of an average firm increases by 44.07% during anti-mafia periods.

We examine whether the European Central Bank (ECB) President’s communication tone is systematically related to market-based measures of euro area bank risk. Using Loughran–McDonald dictionaries, we construct quarterly indicators of textual tone and readability from ECB presidential statements over 2008Q1–2024Q4. We link these indicators to country-level bank credit default swap (CDS) indices for euro area member states. Our analysis focuses on uncertainty-related and constraining language as communication dimensions especially relevant to financial stability. We estimate panel regressions with macro-financial controls, horse-race specifications, and deploy principal component analysis to distinguish overlapping sentiment dimensions, with additional robustness checks in first differences. A more uncertainty-laden and constraining ECB presidential tone is associated with higher bank CDS spreads, even after controlling for macro-financial fundamentals. Lower readability of ECB statements is likewise linked to greater bank stress, indicating that both the tone and the clarity of central bank communication matter for bank-risk pricing. The association between bank CDS spreads and ECB communication tone is more pronounced during turbulent than tranquil periods. Our findings are consistent with a sentiment-guidance channel through which central bank communication is reflected in perceptions of bank fragility beyond traditional policy and macro-financial factors. They indicate that the tone, complexity, and crisis-contingent framing of ECB communication are economically relevant for financial-stability pricing and the sovereign-bank nexus in the euro area.

This study investigates the impact of skilled labor mobility on bank fragility. We construct a novel bank-level measure of skilled labor mobility using job-posting data and show that it significantly increases bank default probability. The effect is driven entirely by demand for risk officers, identifying risk management capacity as the specific mechanism. The effect is amplified by competition and is concentrated among banks experiencing net employee outflows. We further document that legal restrictions on labor mobility weaken the potential detrimental effect. Our findings indicate that skilled labor mobility exacerbates bank fragility, yet targeted policies can bolster resilience against possible adverse effects from labor market frictions.

We develop a credit-risk based framework for macroprudential policy that treats systemic risk as the credit exposure of a policymaker to a portfolio of supervised banks. Using a structural credit model estimated from Credit Default Swap (CDS) prices, we derive the socially optimal capital buffers per bank by balancing the benefits of lower systemic risk against the economic costs of higher capital requirements through reduced credit supply. Applying this framework to Europe’s systemic banks, we find that the market-based optimal buffers are substantially higher than those currently in use and vary significantly across institutions. Conditional on current country averages, within-country buffers are aligned close to optimally; suboptimality mostly arises from substantial differences between country averages.