Payout suspensions during the Covid-19 pandemic

We provide evidence on the unprecedented rate at which firms suspended dividend payments and share repurchases following the outbreak of the Covid-19 pandemic, compare it to the Global Financial Crisis, and estimate the amount of cash firms saved through payout suspensions.


Introduction
The outbreak of the Covid-19 pandemic in the Spring of 2020 caused economic disruption on a scale and at a speed that were unprecedented in modern history. In this environment, attempts at forecasting the magnitude and duration of the pandemic's impact on corporate earnings posed unique challenges, leading many firms to suspend their dividend payments and share repurchase programs. 1 A similar phenomenon, though on a smaller scale, took place during the Global Financial Crisis (GFC) in 2008-09. Both periods experienced high levels of financial market distress and skyrocketing payout suspension rates.
The speed with which economic events unfold and firms respond during financial crises such as the GFC and Covid-19 makes it crucial to examine corporate actions at a higher frequency than is common in the literature on payout decisions. 2 We accomplish this by using daily data to demonstrate how fast firms can act to preserve internal funding at times when the economic value of cash becomes very high.
We find several interesting results. First, the number of US firms that suspended their dividends and buyback programs after the Covid-19 lockdown far exceeded those during the GFC. Second, firms' suspension decisions occurred far more rapidly and were concentrated in a much shorter period of time (mainly March through May of 2020) as compared to the GFC. Third, cash savings from payout suspensions during 2020 were large for many firms and helped preserve liquidity at a critical juncture.
Our analysis is related to recent studies that examine the impact of the pandemic on firms' efforts to raise funds by issuing bonds (Halling et al., 2020) and stocks (Hotchkiss et al., 2020) and by drawing down bank credit lines (Acharya and Steffen, 2020). 3 Our paper is the first to examine how firms' decisions on dividend payouts and buybacks were affected by the Covid crisis and compare this to firm behavior during the GFC. 4 Our analysis proceeds as follows. Section 2 introduces our data while Sections 3 and 4 report our empirical findings on dividend and buyback suspensions. Section 5 concludes.

Data
We begin our analysis by providing details on how we collect daily data on announcements of dividend and share repurchases, including those made by firms that suspended dividends and buybacks. Our data sample starts in 2005 and ends in December 2020, ensuring that we cover the GFC as well as the most intense part of the Covid-19 pandemic; very few payout suspensions were announced after 2020.
Our analysis merges data from several sources. First, using data from the Center for Research in Security Prices (CRSP) from January 2005 through December 2020, we extract daily stock prices, shares outstanding, and dividend announcements for individual firms. This sample includes all ordinary cash dividends declared by US firms with common stocks (share codes 10 and 11) listed on the NYSE, NASDAQ, or AMEX exchanges.
The CRSP data provides detailed information on dividend announcements and distributions but does not include information on dividend suspension dates. To obtain information on dividend suspensions, we rely on two other data sources. First, for each of the public companies in the CRSP data, we use the EDGAR database to download all 8-K forms that companies filed to the SEC between January 2005 and December 2020. 5 Second, we use the NASDAQ news platform to download press releases on companies in our sample. Between January 1 and December 31, 2020, we identify a total of 122,706 press releases with a clear spike around late February and late April.
Combining the textual data from EDGAR and the NASDAQ news platform, we next identify the 8-K filings and press releases that mention dividend suspensions in either the text or title and extract the date of the suspension and the associated ticker using an automated text scraper. This process yields an initial list of 1765 dividend suspensions. After manually reviewing each case to remove false positives, we identify a total of 498 dividend suspensions in our sample.
Similarly, we collect buyback suspension data from a variety of sources. First, we obtain data on suspended and cancelled buybacks from Capital IQ. In addition, we scrape 8-K forms and data on company press releases as we did for the dividend suspensions. We manually check every single buyback suspension 5 In January and February of 2020, the number of 8-K filings was 3225 and 3993, respectively, before rising sharply to 4358 and 4890 in March and April, 2020, and reaching an all-time high of 5812 during May.
date to obtain a final sample of 497 buyback suspensions from 2005 through 2020. 6

Dividend suspensions
The top panel in Fig. 1 shows the total number of dividend suspension announcements aggregated by month between January 2005 and December 2020. Typically less than two or three firms (and often none) announce a suspension of dividends in any given month. The two notable exceptions to this are the GFC (2008GFC ( :01-2009  These numbers are dwarfed by dividend suspensions during the Covid pandemic. In March 2020, 51 firms announced they had suspended their dividends, followed by another 81 announcements in April and nearly 60 in May before suspensions tapered off to 12 in June and returned to normal levels after August 2020. In total, 219 dividend suspensions were announced in 2020.
How much cash did firms actually preserve by suspending or reducing their dividend payments? To address this question, Fig. 2 plots the actual dividend cuts -along with dividend increases -announced by individual firms and summed, each month, across all firms in our sample. We also show the imputed dollar value of dividend suspensions, computed by assuming that the dividend-suspending firms, had they not announced a dividend stop, would have paid the same dividends in a given month as they did in 2019.
This imputed estimate of firms' savings on dividend payments is, as we would expect, zero or extremely small in January and February 2020 and remains quite small in March and April. From May onward, the value rises to a level between $3bn and $4bn in most months. From January through March, the dollar value of dividend rises outpaces the value of any dividend cuts. In May, July, August, and October the two balance out, whereas the dividend cuts are at least twice as large as dividend increases in June and December and much larger in November of 2020. Overall, between March and December of 2020, the firms in our sample saved around $29bn through dividend suspensions. Firms saved substantially more over the same period by cutting dividends by approximately $56.5bn, with the bulk concentrated in November and December of 2020. This figure exceeded the dollar value of dividend rises over the same period ($36.5bn) by $20bn.
These aggregate figures do not reveal the importance of dividend suspensions to individual firms' cash holdings. To shed light on this issue, Fig. 3 provides a histogram of the value of the suspended dividends (assuming these would have been the same as dividends during 2019Q4) as a percentage of holdings of cash and cash equivalent at the individual firm level measured at the end of 2019. Dividend suspensions added 10%-20% to cash holdings for many firms, and amounted to a substantially higher fraction of cash holdings for some individual firms. Notice that these ratios only reflect cash savings from suspending dividends for a single quarter. If, instead, dividends were suspended for, say, four quarters, the ratio should correspondingly be multiplied by four, so dividend suspensions contributed quite substantially to individual firms' cash savings.
Dividend suspensions also varied substantially across industries during the two crises in our sample. The first two columns of Table 1 list the industry composition of dividend suspending firms during 2008-09 and 2020 using the 17-industry classification scheme from Ken French's website. During the GFC, Banks, Insurance Companies and Other Financials counted for nearly half of all dividend suspensions (63 of 135), with Other and Automobiles counting for another 21 and 13 suspensions, respectively. 7 During the Covid-19 pandemic, the Other sector counted 7 The ''Other'' category includes many service firms, e.g., hotels such as Hilton, Marriott, and Choice Hotels and gambling/entertainment/casinos such as Las Vegas Sands and Boyd Gaming Corporation.
for 30% of dividend suspensions (66 of 219) while Retail Stores (31) and Textiles, Apparel and Footwear (11) took up another 20% combined. Conversely, Financial firms (29) counted for less than 15% -a sharp reduction from the GFC. Oil and Petroleum Products, Machinery and Business Equipment, and Automobiles each counted for at least 10 dividend suspensions during the Covid pandemic.

Buyback suspensions
Buyback suspensions (Fig. 1, bottom panel) follow a similar pattern to dividend stops with few cases -typically less than three -announced in normal months. During the GFC, buyback suspensions peak at 15 in October of 2008 before gradually tapering off and reaching normal levels in early 2009. Buyback suspensions picked up far more rapidly and were more concentrated in time during the Covid pandemic; suspensions peak at over 130 in March 2020, followed by 91 in April and another 26 in May. Hence, more firms (248) suspended buybacks than suspended dividends (191) during the turbulent first three months of the pandemic (March-May, 2020). By August, buyback suspension numbers were back to normal.
In total, buyback suspensions tripled in numbers during the pandemic relative to the GFC (262 versus 78). Banks, Insurance Companies and Other Financials, Other, and Machinery and Business Equipment lead the industries with the highest number of buyback suspensions during both crises (columns 3 and 4 in Table 1). In addition, 33 firms in the Retail Stores industry suspended buybacks during the pandemic -far more than during the GFC.
It is difficult to reliably estimate how much money firms saved by suspending their share repurchase programs during the pandemic. Many programs do not commit firms to buy back shares on a particular schedule and some firms might simply have chosen to let their share repurchase programs lapse without formally announcing their suspension. With this important caveat in mind, if we assume that the buyback suspenders would have carried out the same amount of share repurchases during 2020 as they did in 2019, we would have expected an additional $140bn of net buybacks in 2020, including $16bn in April, $22.5bn in July, and $33bn in October.   (2008)(2009) and the Covid-19 crisis (2020), broken down by industry. We use the SIC codes and the Fama-French 17 industry definitions to classify companies into the various industries.

Conclusion
US firms suspended their dividend payments and share repurchase programs in unprecedented numbers and at unparalleled speed after the outbreak of the Covid-19 pandemic. We provide a detailed analysis of the timing of these decisions, compare them to events during the Global Financial Crisis, and evaluate their economic significance by quantifying how much internal cash US firms preserved through payout suspensions.

Data availability
The authors do not have permission to share data.