Big Data Availability and Asymmetric Voluntary Disclosures

We exploit staggered releases of satellite data on parking lot traffic across U.S. retailers and conduct the stacked Poisson regression to identify the causal effect of big data availability on corporate voluntary disclosure. We document a significant decrease in management good news forecasts following the satellite data release, whereas bad news forecasts are unaffected. The asymmetric decrease cannot be explained by the demand-side substitution hypothesis. Our evidence suggests that the supply-side “meeting guidance” hypothesis is likely to account for the disappearing good news forecasts. We also derive cross-sectional predictions from the hypothesis and confirm them in the data: The decrease in good news forecasts is more pronounced among short-horizon quarterly guidance than long-term forecasts and when firms have higher institutional ownership, higher operating uncertainty, and higher litigation risk of missing guidance. This paper was accepted by Suraj Srinivasan, accounting. Funding: S. Wang acknowledges financial support from the National Natural Science Foundation of China [Grant 72373110] and the Fundamental Research Funds for the Central Universities in China. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2023.00816 .