Working Paper: NBER ID: w31936
Authors: Azi Benrephael; Bruce I. Carlin; Zhi Da; Ryan D. Israelsen
Abstract: We use minute-by-minute Bloomberg online status microdata during 2017-2021 to directly study how hard and soft information collection affects equity analyst performance. Collection of hard information, proxied by office workday length, is positively associated with the quantity and timeliness of analyst reports. Soft information collection, as proxied by propensity to travel, is positively correlated with the market’s reaction to recommendation changes and the likelihood of becoming a star analyst. Both hard and soft information collection improve forecast precision, a causal result that we confirm using the COVID lockdown as an instrument.
Keywords: No keywords provided
JEL Codes: D82; D83; G29
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| percentage of away days (PAD) (C23) | market reactions to recommendation changes (G24) |
| percentage of away days (PAD) (C23) | likelihood of becoming a star analyst (G24) |
| average workday length (AWL) (J22) | forecast output (C53) |
| average workday length (AWL) (J22) | forecast precision (C53) |
| percentage of away days (PAD) (C23) | forecast precision (C53) |
| commuting time (R41) | average workday length (AWL) (J22) |