The extant literature predicts market returns with "simple" models that use only a few parameters. Contrary to conventional wisdom, we theoretically prove that simple models severely understate return predictability compared to "complex" models in which the number of parameters exceeds the...

We study disruptions in debt markets during the COVID-19 crisis. The safer end of the credit spectrum experienced significant losses that are hard to fully reconcile with standard default or risk premium channels. Corporate bonds traded at a large discount to their corresponding CDS, and this...

Liquidity provision is a bet against private information: if private information turns out to be higher than expected, liquidity providers lose. Since information generates volatility, and volatility co-moves across assets, liquidity providers have a negative exposure to aggregate volatility...

We build a macroeconomic model that centers on liquidity transformation in the financial sector. Intermediaries maximize liquidity creation by issuing securities that are money-like in normal times but become illiquid in a crash when collateral is scarce. We call this process shadow banking. A...

Managed portfolios that take less risk when volatility is high produce large alphas, substantially increase factor Sharpe ratios, and produce large utility gains for mean-variance investors. We document this for the market, value, momentum, profitability, return on equity, and investment factors...

At the announcement of a new policy, agents form a view of state-contingent policy actions and impact. We develop a method to estimate this state-contingent perception and implement it for many asset-purchase interventions worldwide. Expectations of larger support in bad states--"policy...

We study information acquisition and dynamic withdrawal decisions when a spreading rumor exposes a solvent bank to a run. Uncertainty about the bank's liquidity and potential failure motivates depositors who hear the rumor to acquire additional noisy signals. Depositors with less informative...

We use textual analysis of high-dimensional data from patent documents to create new indicators of technological innovation. We identify significant patents based on textual similarity of a given patent to previous and subsequent work: these patents are distinct from previous work but are...

We perform a comparative analysis of machine learning methods for the canonical problem of empirical asset pricing: measuring asset risk premia. We demonstrate large economic gains to investors using machine learning forecasts, in some cases doubling the performance of leading regression-based...

We study how sophisticated investors, when faced with changes in information environment, adjust their information acquisition and trading behavior, and how these changes in turn affect market efficiency. We find that, after exogenous reductions of analyst coverage due to closures of brokerage...