Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We study the effect of oil news shocks on U.S. aggregate and industry-level stock returns. Using a Proxy-VAR and a sample from January 1973 to December 2019, we find no significant effect on aggregate stock price index on impact, but a persistent and significant drop at longer horizons. An important degree of heterogeneity is found in the industry-level responses: stock returns for precious metals, coal, petroleum and natural gas, and utilities increase significantly and persistently after the shock, whereas consumer goods, rubber and plastic, automobiles, and trucks fall briefly. Moreover, our estimates indicate that oil news shocks pose a risk for IT sectors whose returns exhibit losses. When we extend the sample to December 2022, we uncover a crucial change in the dynamics of the oil surprise measure: it is contaminated by lags of the oil price. We illustrate how using the oil surprise as an instrumental variable in the extended sample produces puzzling responses of industry-level stock returns, whereas a purged measure of the oil surprise leads to more stable estimates.