Working Paper: NBER ID: w29513
Authors: Jennie Bai; Massimo Massa
Abstract: We study information substitutability in the financial market through a quasi-natural experiment: the pandemic-triggered lockdown that has hampered people’s physical interactions and hence the ability to collect, process, and transmit human-interaction-based information. Exploiting the cross-sectional and time-series variation of lockdown and its implications for proximate investment, we investigate how the difficulty of using human-interaction-based information in lockdown has prompted a switch to non-interaction-based information. We show that lockdown reduces fund investment in proximate stocks and generates a portfolio rebalancing toward distant stocks. Such rebalancing negatively impacts fund performance by reducing fund raw (excess) returns an additional 0.51% (0.19%) per month during lockdown, suggesting that human-interaction-based and non-interaction-based information is not easily substitutable. Last, we show that the edge of human-interaction-based information originates preeminently from physical contacts, primarily in cafés, restaurants, bars, and fitness centers, and that the virtual world based on Zoom/Skype/Teams cannot substitute for personal meetings in generating sufficient information.
Keywords: Information substitutability; Human interaction; Lockdown; Investment performance
JEL Codes: G01; G12; G14; G23
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| lockdown (K40) | reduced ability of fund managers to collect and process human-interaction-based information (G41) |
| reduced ability of fund managers to collect and process human-interaction-based information (G41) | decline in investment in proximate stocks (G31) |
| decline in investment in proximate stocks (G31) | shift toward distant stocks (G19) |
| shift toward distant stocks (G19) | negative impact on fund performance (G23) |
| inability to substitute human-interaction-based information with non-human-interaction-based information (D83) | more passive investment strategy among local-investing funds during lockdown (G23) |
| more passive investment strategy among local-investing funds during lockdown (G23) | decrease in investments in proximate stocks (G11) |
| more passive investment strategy among local-investing funds during lockdown (G23) | reduce excess weight deviated from benchmark index (C43) |
| reduction in social interactions (Z13) | no significant effect on behavioral biases of familiarity and trust (G41) |