Abstract
This thesis examines how U.S. firms adjusted dividend policies during the Global Financial Crisis (2008–2009) and the COVID-19 pandemic (2020–2021). Dividend event data from CRSP were linked to Compustat fundamentals, and three models were estimated: a logit regression for dividend cuts, a linear fixed-effects model for payout intensity, and a difference-in-differences approach to test leverage heterogeneity. The results show that dividend cuts were significantly more likely during the GFC, consistent with credit distress, while COVID generated a weaker and more heterogeneous effect. Firm fundamentals remained decisive: leverage increased both the probability of cuts and the retention ratio, while larger firms displayed resilience. Profitability showed mixed results, positively linked to cuts in the logit model but negatively related to payout in the linear model, suggesting reinvestment motives alongside distributional choices. The analysis also revisits dividend signaling. In the GFC, maintaining payouts was costly and thus highly informative, while cuts were interpreted as strong negative signals. During COVID, widespread suspensions diluted informational content, and stability or increases became the true signals of resilience.| Date of Award | 17 Oct 2025 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | José D. Garcia Revelo (Supervisor) |
UN SDGs
This student thesis contributes to the following UN Sustainable Development Goals (SDGs)
-
SDG 8 Decent Work and Economic Growth
-
SDG 9 Industry, Innovation, and Infrastructure
Keywords
- Dividend policy
- Signaling
- Global financial crisis
- COVID-19
- Payout behavior
Designation
- Mestrado em Finanças (mestrado internacional)
Cite this
- Standard