Real earnings manipulation and future performance: A revisit using quarterly data of firms with debt covenants

• Date: 01 January 2020
• Author: Kenneth Zheng,Weiwei Wang
• Published date: 01 January 2020
• DOI: http://doi.org/10.1002/rfe.1070

76

|

wileyonlinelibrary.com/journal/rfe Rev Financ Econ. 2020;38:76–96.

© 2019 University of New Orleans

1

|

INTRODUCTION

Extant literature has provided much evidence on real earnings manipulation (hereafter, REM). For instance, managers may

decrease the amount of discretionary SG&A expenditures or increase production in order to lower the costs and boost earnings

(e.g., Roychowdhury, 2006; Bens, Nagar, & Wong, 2002). REM differs from accruals earnings management (hereafter, AEM)

in that REM affects cash flows by altering business operations whereas AEM influences earnings through the timing of the ac-

cruals but has no effect on cash flows. Additionally, REM is more difficult to be detected than AEM. Even if REM is detected,

the auditors are unlikely to question the modified activity levels (Graham, Harvey, & Rajgopal, 2005). Although managers may

use a combination of REM and AEM to manage earnings, they tend to use REM during the fiscal year and AEM close to or

after year‐end (Zang, 2012).

A recent line of literature investigates the implications of REM on firms’ operating performance and firm value. Most

studies in this literature document a negative relationship between REM and firm performance (e.g., Cohen & Zarowin, 2010;

Eldenburg, Gunny, Hee, & Soderstrom, 2011; Mizik, 2010). However, Gunny (2010) finds that those firms that use REM to just

meet earnings targets have relatively better future performance than firms that do not use REM and miss or just meet earnings

benchmarks. Furthermore, Vorst (2016) finds that REM that reverses in the year following an abnormal reduction in invest-

ments is associated with lower future performance, suggesting that reversing REM is indicative of true REM.

Received: 24 April 2019

|

Revised: 3 June 2019

|

Accepted: 17 June 2019

DOI: 10.1002/rfe.1070

ORIGINAL ARTICLE

Real earnings manipulation and future performance: A revisit

using quarterly data of firms with debt covenants

WeiweiWang1

|

KennethZheng2

1School of Accounting and Taxation,Weber

State University, Ogden, Utah

2Department of Accounting and Finance,

College of Business,University of

Wyoming, Laramie, Wyoming

Correspondence

Weiwei Wang, School of Accounting and

Taxation, Weber State University, 1,337

Edvalson Street, Dept. 3,808, Ogden, UT

84408–3803.

Email: weiweiwang@weber.edu

Abstract

We investigate the implications of real earnings manipulation (REM) and reversals

of REM on firms’ future operating performance using quarterly data of firms with

debt covenants. In the presence of debt covenants, firms are under persistent pressure

to deliver financial results that exceed the thresholds of the debt covenant require-

ments. We find that REM is associated with lower future operating performance.

More importantly, the reversals of REM in the following quarter have an incremental

positive effect on future performance, which largely offsets the negative effect of

REM. These results provide new evidence on REM reversals that differs from the

existing literature. Instead of interpreting the reversals as an indication of true REM

based on their negative association with future performance documented in Vorst

(2016), our results suggest that REM reversals may be indicative of firms rewinding

REM subsequently, which reduces the REM damage to firms’ future operations.

KEYWORDS

debt covenant, operating performance, real earnings manipulation

JEL CLASSIFICATION

D24; M11; M30; M41

|

77

WANG ANd ZHENG

We extend the literature by examining the relationship between REM and firms’ future operating performance using a

sample of quarterly data of firms with debt covenants. This sample is distinct from that used by Vorst (2016) in two important

aspects. First, firms with debt covenants are under consistent pressure on a quarterly basis to deliver financial results that meet

or exceed the debt covenant requirements. Prior studies have shown that technical violations of debt covenants occur frequently

(e.g., Watts & Zimmerman, 1986; DeAngelo, DeAngelo, & Skinner, 1994; Dichev & Skinner, 2002) and that firms manage

earnings to avoid violations of debt covenants (Efendi, Srivastava, & Swanson, 2007; Franz, HassabElnaby, & Lobo, 2014;

Jaggi & Lee, 2002; Jha, 2013; Sweeney, 1994). Given that quarterly firm performance is affected by seasonality in business,

it is vital for firms to manage quarterly earnings to satisfy debt covenants. Second, based on the findings in Zang (2012, firms

engage in more REM during the year and more AEM close to or after year‐end because managers adjust the latter type of earn-

ings management based on the outcome of the former. Since our sample consists of quarterly data, firms’ use of REM is likely

more salient than that of AEM in our sample compared to the Vorst (2016) sample based on annual data.

Due to these characteristics of our sample, the implications of reversing REM in the context of persistently meeting or ex-

ceeding the debt covenant thresholds on a quarterly basis may differ from those of reversing REM on an annual basis in Vorst

(2016). Specifically, Vorst (2016) find that REM that reverses in the following year is, on average, a strong precursor of poorer

future industry‐adjusted performance. In our context, however, if a firm chooses to conduct REM in a quarter to comply with

the debt covenant requirements, a reversal of the REM in the following quarter should indicate that the financial results in the

following quarter are strong enough for the firm to unwind the REM employed in the previous quarter. To the contrary, if the

firm does not reverse the quarter t REM in t+1, then it may suggest that the firm continues to operate under the pressure of

meeting the debt covenant requirements. As the horizon of REM lengthens, the cost of REM on the firm's operations is likely

to increase.

In this study, we measure REM by three dummy variables. First, we create a dummy variable that equals one if a firm‐

quarter has abnormally low SG&A expenses, and zero otherwise. Second, we generate a dummy variable that equals one if a

firm‐quarter has abnormally high production costs, and zero otherwise. Third, we form a dummy variable that equals one if the

aggregate value of abnormal SG&A minus abnormal production cost for a firm‐quarter is abnormally low, and zero otherwise.

Using a sample of 94,589 firm‐quarter observations of firms with debt covenants in their loan contracts spanning from 1992 to

2016, we provide evidence that REM in quarter t is associated with lower industry‐adjusted earnings or industry‐adjusted cash

flow from operations in quarters t+2 and t+3. More importantly, we show that REM reversals in quarter t+1 have an incre-

mental positive effect on firms’ operating performance in quarters t+2 and t+3. Our finding suggests that the implications of

reversing REM in the context of persistent pressure to exceed the debt covenant requirements differ from those of Vorst (2016)

in the absence of this pressure.

In additional analyses, we examine the effect of REM and REM reversals using a subsample of firm‐quarter observations

consisting of firms that are close to violation or in technical default of either the current ratio debt covenants or the net worth

debt covenants. The results confirm our main finding that in the presence of the pressure to meet the debt covenant require-

ments, REM negatively affects future operating performance. Moreover, firms that reverse the quarter t REM in t+1 are able

to largely offset the negative effect of REM on future performance.

The main contribution of this study is that reversing REM has differing implications on future performance conditional on

the purpose and time horizon of REM. Our findings extend the Vorst (2016) finding by documenting that REM reversals do

not indicate that the negative future performance consequences are stronger under certain circumstances. Given a frequent need

to meet the debt covenant thresholds, an REM reversal is an indicator of incremental positive future performance that largely

offsets the negative effects of REM. Additionally, our results contribute to the literature on REM. Graham et al. (2005) find that

managers are willing to take actions that have an adverse real economic effect in order to meet earnings targets. Our results add to

this line of literature by showing that when firms with financial covenants employ REM to meet the debt covenant requirements

in the short run and are able to reverse it quickly, the negative economic effect on future performance significantly reduces.

The rest of the paper proceeds as follows. Section 2 reviews the literature and develops the hypothesis. Section 3 discusses

the methodology. Section 4 provides the empirical results and section 5 concludes.

2

|

LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT

2.1

|

Literature review

Recent research has shown that firms have incentives to avoid debt covenants violations because covenant violations may cause

severe economic consequences. For instance, firms’ capital investment may decline sharply following a financial covenant

violation as creditors may impose the threat of accelerating the loan to intervene in management (Chava & Roberts, 2008).