In this seminar Benjamin Wong proposed a way to directly nowcast the output gap using the Beveridge-Nelson decomposition based on a mixed-frequency Bayesian VAR. The mixed-frequency approach produces very similar estimates of the U.S. output gap to those based on a quarterly model, but it is able to provide timely updates to estimates within a quarter as higher-frequency data become available. The author found that the output gap nowcasts are much more reliable than those for output growth, with monthly indicators for a credit risk spread, consumer sentiment, and initial claims containing particularly useful information about the output gap. An out-of-sample analysis of the COVID-19 crisis anticipated huge declines in the U.S. output gap in the first half of 2020 before quarterly real GDP was observed, with a scenario nowcast given data up to April 2020 implying an output gap of -28.8% in 2020Q2.