Working Paper: NBER ID: w27829
Authors: Harrison Hong; Jeffrey D. Kubik; Neng Wang; Xiao Xu; Jinqiang Yang
Abstract: We derive a parsimonious model of damage to corporate earnings from COVID-19. Using measures of expected damage from industry-level earnings forecast revisions, we estimate this model with nonlinear least squares and identifying restrictions related to forecast rationality. Forecasts in mid-May 2020 imply an earnings crash and lower earnings growth until a vaccine arrives in 1.48 years (95% CI [0.61, 5.88]). We extend our framework to account for time-varying vaccine arrival rates. Mid-August 2020 forecasts imply a vaccine arrival in 0.61 years (95% CI [0.35, 1.06]), which is due to positive vaccine news as opposed to fiscal or monetary policy news.
Keywords: COVID-19; Earnings Damage; Vaccine Arrival; Forecast Revisions
JEL Codes: G10; G12; G31; G32
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| vaccine arrival expectations (I19) | earnings forecasts (G17) |
| vaccine arrival (I19) | corporate earnings recovery (G38) |
| pandemic (F44) | lower earnings growth (J31) |
| vaccine arrival (I19) | jump in earnings (J31) |
| earnings crash (G01) | lower earnings growth until vaccine arrival (J31) |