IMPLICATIONS OF UNSUSTAINABLE DIVIDENDS

• Date: 01 March 2020
• Author: Oneil Harris,Jurica Susnjara,Thanh Ngo
• DOI: http://doi.org/10.1111/jfir.12204
• Published date: 01 March 2020

The Journal of Financial Research Vol. XLIII, No. 1 Pages 185–225 Spring 2020

DOI: 10.1111/jfir.12204

IMPLICATIONS OF UNSUSTAINABLE DIVIDENDS

Oneil Harris and Thanh Ngo

East Carolina University

Jurica Susnjara

Barry University

Abstract

In this article, we examine the wealth effects of unsustainable dividend payments and

explore the economic forces that may explain why they exist. We find that the larger

the dividend–earnings differential, the lower the short‐and long‐run wealth effects to

shareholders. In addition, the dividend–earnings differential increases not only the

probability of a subsequent dividend cut over the next four quarters but also the

likelihood that the cut will be greater than 5%. Overall, our findings suggest that

although investors are not fooled by unsustainable dividend payments, the negative

announcement effects are in anticipation of protracted poor performance.

JEL Classification: G32, G35

I. Introduction

Research indicates that managers are generally reluctant to cut dividends and willing to

take costly actions to avoid a cut (see Allen and Michaely 2003; Daniel, Denis, and

Naveen 2010; Brav et al. 2005; DeAngelo and DeAngelo 1990; DeAngelo, DeAngelo,

and Skinner 2008). Brav et al. (2005) conduct in‐depth interviews with financial

executives and note that managers consider the importance of dividend stability to be

on par with investment decisions. They report that managers express a strong aversion

to reducing dividends, except in extraordinary circumstances. For example, rather than

cutting dividends, some executives convey a willingness to pass up positive net present

value projects; others are more eager to sell assets, raise external funds, or lay off

employees before considering a dividend reduction. Likewise, Daniel, Denis, and

Naveen (2010) find that firms rarely respond to cash‐flow shortfalls by cutting

dividends. Instead, the vast majority borrow more heavily and commonly cut back on

their investment levels.

Jensen, Lundstrum, and Miller (2010) point out that executives’reluctance to

cut dividends emanates from the view that dividend reductions signal managerial

pessimism about future earnings, which is supported by the significant negative market

reaction to dividend cuts (e.g., Grullon, Michaely, and Swaminathan 2002; Allen and

Michaely 2003). As a result, managers deem a dividend reduction to be a “last resort”

(Jensen, Lundstrum, and Miller 2010). Kaplan and Reishus (1990), Fudenburg and

Tirole (1995), and Parrino, Sias, and Starks (2003) suggest that the reluctance is partly

185

© 2019 The Southern Finance Association and the Southwestern Finance Association

motivated by managers’career concerns, as a dividend cut reduces the chance that an

executive retains her current position or obtains a future directorship.

Ghosh (1993) offers an explanation for the preceding phenomena based on

regret theory a nd develops a mode l that encapsulates managerial reluctance to c ut

dividends. Regret theory implies that managers, when making decisions on uncertain

prospects, experience pride or regret about their actions depending on the outcome (Bell

1982; Fishburn 1982; Loomes and Sugden 1982). In line with this view, Ghosh reports

that in making dividend decisions, executives exhibit risk aversion in choices over gains;

in choices over a loss, they exhibit risk‐seeking behavior. He surveys executives and

finds that (1) in a choice between paying higher dividends or conserving funds for

uncertain investment opportunities, managers prefer higher dividends (risk aversion), and

(2) in a choice between cutting dividends or supporting dividends with funds borrowed at

high cost and risk, managers prefer to support dividends (risk proneness). Two notable

predictions emerge from this behavioral perspective: (1) the simultaneous payment of

dividends and sale of securities to finance investments and (2) the continuation of

dividends using external funds when earnings are inadequate to support them. Based on

regret theory, Ghosh affirms that the decision to continue with dividends when earnings

are declining is motivated by pride.

Our inquiry is inspired by this germinal work that describes how executives

think about dividends when making decisions under uncertainty and, in particular, the

prediction that managers try to avoid dividend cuts even in the face of poor earnings.

Anecdotal evidence suggests that several companies recently paid out more in

dividends in a given year than they generated in total earnings. USA TODAY, for

instance, reports in a 2016 article that 42 companies in the S&P 500 index paid

dividends in the preceding 12 months that exceeded their reported net income.

1

Using a

broader cross‐section of companies, we provide evidence showing that about one in

every five quarterly dividend payments is, in fact, greater than stated earnings, and this

estimate excludes firms with negative earnings. Thus, the practice of paying a larger

dividend than reported earnings is more widespread than previously realized. Yet, the

valuation consequences of paying these types of high‐risk dividends remain unclear.

Under regret theory, managers’persistence in paying cash dividends during

periods of poor earnings performance is motivated by the belief that if prospects

improve, they can proudly take credit for steering the firm out of a crisis while

avoiding a dividend cut and its resultant capital loss. However, if prospects fail to

improve in the near term, shareholders could suffer more devastating losses (e.g., in a

bankruptcy or reorganization) relative to the loss of a dividend cut. As a result,

investors may view these kinds of dividend payments unfavorably. The literature has

paid scant attention to the valuation impacts of continuing with dividends despite cash‐

flow problems. As a result, little is known about the market’s perception of these types

of high‐risk dividend payments or their associated long‐horizon price effects.

1

Matt Krantz, “These Companies Pay Out More Than They Earn,”USA TODAY (July 7, 2016), https://

www.usatoday.com/story/money/markets/2016/07/07/these‐companies‐pay‐out‐more‐than‐they‐earn/

86778320/.

186 The Journal of Financial Research

Moreover, the Ghosh (1993) study is limited to a survey of only 250 executives more

than two decades ago, and he does not address the valuation impacts of these high‐risk

payouts.

Our article extends prior research by taking a detailed analytical look at the

wealth effects of unsustainable dividend payments and the plausible economic forces

that may explain why, and through what channels, these disbursements operate. We

define unsustainable dividend payers as firms in which corporate payout exceeds stated

earnings. For robustness, we measure both payout and earnings in diverse ways. More

specifically, we compute corporate payout using both the cash dividend amount and the

total payout, where the total payout is the summation of dividends and repurchases. We

employ four alternative instruments to identify an unsustainable dividend payer: (1)

where dividend per share (DPS) exceeds stated earnings per share (EPS), (2) where

total dollar dividend amount exceeds operating income, (3) where total payout exceeds

operating income, and (4) where total payout exceeds net income.

We predict that investors will subject unsustainable dividend payers to a larger

risk premium than other dividend‐paying firms, as their high‐risk payouts expose

shareholders to potentially larger losses compared to the loss associated with a

dividend cut. Hence, we expect unsustainable dividend payments to elicit unfavorable

short‐and long‐run abnormal returns. We find that the stock price reaction to

unsustainable dividend payments is negative and the three‐day cumulative abnormal

returns (CARs) are considerably lower than that of other dividend payment

announcements. As anticipated, we further find that the larger the differences between

the dividend and EPS, the worse the announcement CAR to shareholders even after

controlling for other factors. The marginal effect of the dividend–earnings differential

on the three‐day CAR is −0.065, which implies that the market value of the average

firm decreases by about $289 million in economic terms around these types of high‐

risk payouts.

2

Unsustainable dividend payers also register significantly lower 12‐month buy‐

and‐hold abnormal returns (BHARs) on a risk‐adjusted basis compared to their peers.

The adverse long‐term price movements imply that unsustainable dividend payments

have a protracted negative effect on firm value. We also find that abnormal long‐

horizon returns are lower when the dividend–earnings differential is larger, which

offers evidence that the adverse announcement CARs are partially in anticipation of

weak long‐term stock price performance following the dividend distribution. We

perform various robustness tests, and this finding continues to hold. The sustained

adverse price response to these high‐risk payouts is in line with the evidence of a post‐

dividend‐announcement drift (see Bae 1996; Howe, Vogt, and He 2003; Michaely,

Thaler, and Womack 1995). Our findings are also consistent with studies showing that

payout reduces the information asymmetry around firm value (see Turki 2019).

2

The $289 million is estimated by multiplying the estimated coefficient of the DPS_EPSPIQ variable in

Panel A of Table 4 (−0.065) by the market value of the average firm in the sample given in Table 2 ($4,443.67

million).

187Implications of Unsustainable Dividends