How could analysts’ facial trustworthiness affect their forecasting accuracy?

Purpose To investigate how the facial trustworthiness of financial analysts affects their ability to privately gather information from corporate insiders. The channels are hypothesized to be managerial features, information environment and legal enforcement. Design/methodology/approach The study employs a quantitative research design using a comprehensive sample of Chinese financial analyst reports. Machine learning techniques for facial analysis are applied to assess the trustworthiness of analysts, and the relationship between facial trustworthiness and the informativeness of the analysts' reports is analyzed. Findings We first find that more trustful-looking analysts can produce more informative reports upon their visits to the corporate insiders. The effect of analysts’ facial trustworthiness is driven by the gender difference and the level of professional competency of the visited insiders. Moreover, when information disclosure of the reported firm is lower, the trust effects become more pronounced. Finally, the trust effect disappears after a legal action against an insider trading activity related to one of the analysts’ visits. Originality/value We find that more trustful-looking analysts can gain more material information during their visits to the firms, particularly when interacting with managers who have lower professional competency and a greater tendency to trust others. This trust-based informational advantage is also amplified in firms with poorer information environments. Furthermore, this trust effect disappears after a legal action against an insider trading activity related to one of the analyst visits. Our results are robust after controlling for analysts’ competency, facial attractiveness and industrial concentration.