Knowledge that Transforms

We examine how different forms of government intervention - explicit deposit insurance, implicit guarantees from nationalization, and assurance effects from re-privatization - shape depositor behavior during a banking crisis. Using proprietary monthly account-level data, we show that depositors respond to formal insurance limits even in an environment of broad state support, but adjust only partially. Stabilization is markedly stronger after the takeover by a financially strong private bank than during state ownership, despite unchanged statutory coverage. These patterns are inconsistent with a strong bailout-expectations view in which explicit insurance is largely irrelevant and instead point to credibility as a key determinant of depositor behavior. We further show that political trust amplifies the effectiveness of government guarantees. Overall, our findings highlight that the form and credibility of government backstops, rather than coverage levels alone, are central to retail deposit funding stability during crises.

An asset owner designs an asset-backed security and a signal about its value. After experiencing a liquidity shock and privately observing the signal, he sells the security to imperfectly competitive buyers. With flexible information design, the security’s informational sensitivity is a blessing, not a curse. Among double-monotone securities, the issuer prefers to sell the asset in combination with an appropriate signal rather than design more complex securities. Within the class of monotone securities, live-or-die securities are optimal and strictly dominate asset sales. When signals are restricted to be sufficiently informative about the upside, debt and additional-tier-1 debt become optimal.

We collect 2,848 instrumental variable (IV) regressions from 391 papers published in top economics and finance journals and show that estimated IV coefficients are often much larger than their OLS counterparts, even when theory suggests they should be smaller. We derive meta-regression estimators that distinguish measurement error bias from omitted variable bias and simultaneity bias; applying our estimators to a set of 1,040 comparable regressions from 139 papers, we find that measurement error shrinks reported coefficients by about two to four times. Measurement error remains quantitatively important even after accounting for weak instruments, publication bias, instrument invalidity, and heterogeneous treatment effects. Our results suggest that measurement error is a severe, pervasive, and understated source of bias in economics and finance.

Prior studies on cryptocurrency momentum overlook liquidation risk and other real-world frictions and inadequately assess its profitability. We evaluate time-series and cross-sectional momentum addressing these issues. When portfolios are marked to market daily, many cross-sectional momentum portfolios are liquidated, and those with statistically significant mean returns often earn negative profits. The mean return is insufficient for testing profitability in the cryptocurrency market, where returns are heavily skewed and fat-tailed. Evidence of time-series momentum is strong, whereas evidence of cross-sectional momentum is almost nonexistent. Investor overconfidence and the disposition effect appear to be important drivers of cryptocurrencies’ price dynamics.