13 Dec 2022
The latest in the Sasin Research Seminar series was a fascinating talk by Saranyu Ungpakorn, Sasin Ph.D. student, on the effects of the economic uncertainty brought about by infectious diseases on corporate dividend policy.
The lecture began with a look at global pandemics dating from 1900 to the present day, including the Spanish Flu, HIV/AIDS, Yellow Fever, Swine Flu, Ebola, and Covid-19. These outbreaks were then added to a chart of World GDP for the same period. This revealed that there didn’t seem to be much correlation between GDP and pandemics, although other significant global events like war affected the data.
Pandemics pose a serious impediment to economic development. For individuals and households, there is the risk of infection and increased cost of the social insurance mechanism. On the other hand, when companies face with the pandemic dissemination, they shall prepare themselves on the unprecedent situations. Some companies brutally execute the early retirement policy with their employee, but some companies that cannot adjust themselves finally collapse down. They will, eventually, experience a loss of production and a rise in medical and death-related benefit payments. Finally, from a macroeconomic point of view, there will be a decline in growth from decreasing populations and an increase in people saving money.
Next was a look at the literature relating to whether a rise in uncertainty and perception of risk would lead to an adjustment of a firm’s dividend policy. The paper also applies agency theory to examine the topic in more detail. This is the concept that when an opportunity arises, the agent – the manager – will pursue personal interests rather than the interests of the principal.
Saranyu then presented the paper’s three hypotheses and the literature that supported them. The first was that firms are prone to decreasing dividends when there is rising uncertainty resulting from infectious diseases. The second hypothesis was that firms would raise dividend distribution in response to a pandemic. Finally, there was the idea that firms would raise dividends if a firm had a strong independent board, regardless of uncertainty due to a disease outbreak.
Despite there being similar studies about uncertainty and businesses, there is a gap in the literature relating to pandemic outbreaks and their effect on dividend payments, which this paper will address.
The talk then moved on to sample selection and the empirical model used. The study is based on a large sample of nearly 288,000 observations on daily basis across four decades from 1985 to 2021, using firms based in the United States.
The dependent and control variable was dividend payments. The independent variable is the Daily Infectious Disease Equity Market Volatility Index, which was then explained in more detail along with the categories measured. The empirical model, which includes the Infectious Disease Index and the variables, was then described.
The results were then presented in a table and discussed. The results show that a higher level of uncertainty induced by infectious diseases diminishes dividends significantly. Specifically, a rise in uncertainty by one standard deviation results in a decline in dividends by almost 2%.
However, the research also revealed that more independent directors on the board mitigated the negative effect of infectious diseases on dividends. This implies that the dividend cut in the presence of infectious diseases is partly motivated by agency conflicts.
Infectious diseases are likely exogenous to firm-specific characteristics because they are beyond any company’s control. As a result, the study’s findings are unlikely tainted by endogeneity and probably reflect a causal influence rather than a mere association.
The presentation was followed by a Q&A session covering various topics and suggestions.
One question asked about the dependent variable and the correlation between infectious diseases, times periods and the consequent influence on dividends. Another asked why firms might increase dividend payments during a crisis. Finally, it was suggested that an interesting take would be to look at the data excluding the effects of Covid-19.