Dividends and share repurchases during the COVID‐19 economic crisis

• Published date: 01 June 2023
• Author: Mieszko Mazur,Man Dang,Thi Thuy Anh Vo
• Date: 01 June 2023
• DOI: http://doi.org/10.1111/jfir.12324

Received: 6 July 2021

|

Accepted: 22 March 2023

DOI: 10.1111/jfir.12324

ORIGINAL ARTICLE

Dividends and share repurchases during the

COVID‐19 economic crisis

Mieszko Mazur

1

|Man Dang

2

|Thi Thuy Anh Vo

2

1

Finance, Accounting, and Management

Control, ESSCA School of Management,

Boulogne‐Billancourt, France

2

Faculty of Finance, University of Economics,

University of Da Nang, Da Nang, Vietnam

Correspondence

Mieszko Mazur, Finance, Accounting, and

Management Control, ESSCA School of

Management, 55 Quai Alphonse le Gallo,

92513 Boulogne‐Billancourt, France.

Email: Mieszko.MAZUR@essca.fr

Funding information

Vietnam National Foundation for Science and

Technology Development (NAFOSTED),

Grant/Award Number: 502.02‐2018.304

Abstract

In this article, we examine dividends and share repurchases

of S&P 1500 firms during the COVID‐19 crisis character-

ized by the stock market crash and a relatively quick stock

price recovery propelled by technology stocks. We find

that the great majority of firms either maintain or increase

the level of dividends during the crisis period. Yet, the

relation between the dividend payout and reported

earnings is negative and significant. This relation also holds

for other types of payouts, including share repurchases and

special dividends. Moreover, we find that both forecasted

and realized earnings of up to 1 year into the future are

negatively associated with current dividends, implying that

existing payout policies are unsustainable in the longer

term. Surprisingly, the difference‐in‐differences test shows

that firms strongly affected by the COVID‐19 crisis have

higher dividend payouts (relative to net earnings) compared

to unaffected firms. The same test indicates that strongly

affected firms significantly reduce repurchases.

JEL CLASSIFICATION

G30, G32, G35

J Financ Res. 2023;46:291–314. wileyonlinelibrary.com/journal/JFIR

|

291

This is an open access article under the terms of the Creative Commons Attribution‐NonCommercial‐NoDerivs License, which

permits use and distribution in any medium, provided the original work is properly cited, the use is non‐commercial and no

modifications or adaptations are made.

© 2023 The Authors. Journal of Financial Research published by Wiley Periodicals LLC on behalf of The Southern Finance

Association and the Southwestern Finance Association.

1|INTRODUCTION

In March 2020, the worldwide COVID‐19 pandemic led all major industrialized countries to shut down the bulk of

their economic activities. Manufacturing and service firms were closed, air traffic was suspended, and populations

were locked down. As a consequence, nations' economies suffered a significant decline in total output and an

accompanying reduction in consumption. The resulting revenue shock led to stock market crashes, gross domestic

product (GDP) contractions, and rising unemployment on an unprecedented scale. In the second quarter of 2020

(2020q2), US GDP fell by 9.5% and its unemployment rate rose to over 20%.

1

In this article, we examine corporate dividend policies during the initial stages of the COVID‐19 crisis. Although

the US stock market followed a V‐shaped recovery, the rebound in stock prices was largely attributable to

technology stocks, whereas other sectors lagged behind their pre‐COVID‐19 capitalizations (Langley, 2020). With

this in mind, we study the dividend behavior of US firms in response to the economywide negative shock induced

by COVID‐19 and its heterogenous impact on revenues and earnings. Similar to other studies (e.g., Campello

et al., 2010), our objective is to provide an early analysis of the effect of the crisis on the real decisions made by

listed firms. We address two basic questions: Can the dividend policy under the COVID‐19 crisis be explained by

prudent management behavior (e.g., DeAngelo et al., 2004; Ham et al., 2020; Lintner, 1956)? Or, alternatively, is the

dividend policy predominantly driven by the agency problems of corporate insiders and their career concerns (e.g.,

Acharya et al., 2017; Fudenberg & Tirole, 1995; Wu, 2018)?

We establish several empirical facts related to the corporate dividend payment decisions in times of crisis. First,

we find that the great majority of US listed firms (82%) either maintain or increase the level of dividends despite the

painful economic downturn. Moreover, the incidence of dividend cuts during the COVID‐19 crisis is 6%, which is

similar to the percentage reported in the normal (noncrisis) periods (see, e.g., Li & Lie, 2006). Furthermore, we find a

negative and significant relation between realized earnings and dividend payout, meaning that the poorest

performers pay out the highest dividends (relative to net earnings). This effect remains robust with respect to the

two most standard measures of dividend policy and is statistically significant at the 1% level throughout all tests.

Next, we examine other types of payouts including repurchases and special dividends. We find that

repurchases are much more sensitive to the crisis than dividends, whereas special dividends are largely random. The

aggregate value of share repurchases declines from $177 billion in 2020q1 to $31 billion in 2020q2 (an 82%

decline). In comparison, the aggregate value of dividends is $140 billion in 2020q1 and declines to barely $131

billion in 2020q2 (a 6% decline). Intriguing, all types of payouts peak in 2020q1; however, most prudent companies

cut or suspend dividends immediately after the imposition of lockdowns in March 2020. When we investigate the

relation between share repurchases and earnings, we find that companies with poor earnings spend heavily on

share repurchases. We find the same pattern for total payouts, which include quarterly and special dividends as well

as share repurchases. Overall, our econometric analysis yields consistent results for all types of payout policies:

Poor performers with the lowest earnings display the highest relative payouts.

In the following analysis, we test the predictive power of the dividend yield (Kothari & Shanken, 1997; Maio &

Santa‐Clara, 2015). Because dividends and stock prices reflect earnings potential, predictability of stock returns

based on dividend yield must be ultimately linked to company earnings. We test this conjecture by examining the

relation between the current dividend yield and future earnings (both expected and realized) 4 quarters into the

future. We find that companies with the highest dividend yield generate the lowest future earnings. Similarly, we

find that high dividend yield is negatively related to analyst consensus forecasts. We obtain qualitatively similar

results when we use the dividend payout ratio.

In the final test, we use the difference‐in‐differences framework and establish that, all else equal, companies

from sectors strongly affected by COVID‐19 have significantly higher dividend payout ratios than companies from

1

https://faculty.fuqua.duke.edu/%7Echarvey/Audio/COVID/COVID-Harvey.html

292

|

JOURNAL OF FINANCIAL RESEARCH