This paper specifically investigates the effects of US government emergency actions on the investor sentiment–financial institution stock returns relationship. Despite attempts by many studies, the literature still provides no answers concerning this nexus. Using a new firm-specific Twitter investor sentiment (TS) metric and performing a panel smooth transition regression for daily data on 66 S&P 500 financial institutions from January 1 to December 31, 2020, we find that TS acts asymmetrically, nonlinearly, and time varyingly according to the pandemic situation and US states’ responses to COVID-19. In other words, we uncover the nexus between TS and financial institution stock returns and determine that it changes with US states’ reactions to COVID-19. With a permissive government response (the first regime), TS does not impact financial institution stock returns; however, when moving to a strict government response (the overall government response index exceeds the 63.59 threshold), this positive effect becomes significant in the second regime. Moreover, the results show that the slope of the transition function is high, indicating an abrupt rather than a smooth transition between the first and second regimes. The results are robust and have important policy implications for policymakers, investment analysts, and portfolio managers.

中文翻译：

---

美国各州对 COVID-19 的反应能否恢复投资者情绪？来自标准普尔 500 强金融机构的证据

本文专门研究了美国政府紧急行动对投资者情绪与金融机构股票回报关系的影响。尽管许多研究都做出了尝试，但文献仍然没有提供有关这种关系的答案。使用新的公司特定 Twitter 投资者情绪 (TS) 指标，并对 2020 年 1 月 1 日至 12 月 31 日期间 66 家标准普尔 500 金融机构的每日数据进行面板平滑过渡回归，我们发现 TS 表现出不对称、非线性和随时间变化的行为根据大流行情况和美国各州对 COVID-19 的反应。换句话说，我们揭示了 TS 和金融机构股票回报之间的联系，并确定它随着美国各州对 COVID-19 的反应而变化。在政府的宽松回应下（第一种制度），TS 不会影响金融机构的股票回报；然而，当转向严格的政府应对（政府总体应对指数超过63.59阈值）时，这种积极影响在第二种制度中变得显着。此外，结果表明，过渡函数的斜率很高，表明第一和第二状态之间的过渡是突然的，而不是平滑的。结果很稳健，对政策制定者、投资分析师和投资组合经理具有重要的政策影响。