The Impact of CFOs? Incentives and Earnings Management
Ethics on their Financial Reporting Decisions: The Mediating
Role of Moral Disengagement

Cathy A. Beaudoin ? Anna M. Cianci ?

George T. Tsakumis

Received: 23 August 2012 / Accepted: 12 February 2014 / Published online: 7 March 2014

? Springer Science+Business Media Dordrecht 2014

Abstract Despite regulatory reforms aimed at inhibiting

aggressive financial reporting, earnings management per-

sists and continues to concern practitioners, regulators, and

standard setters. To provide insight into this practice and

how to mitigate it, we conduct an experiment to examine

the impact of two independent variables on CFOs? dis-

cretionary expense accruals. One independent variable,

incentive conflict, is manipulated at two levels (present and

absent)?i.e., the presence or absence of a personal finan-

cial incentive that conflicts with a corporate financial

incentive. The other independent variable is CFOs? earn-

ings management ethics (??EM-Ethics,?? high vs. low),

measured as their assessment of the ethicalness of key

earnings management motivations. We find that incentive

conflict and EM-Ethics interact to determine CFOs? dis-

cretionary accruals such that (a) in the presence of incen-

tive conflict, CFOs with low (high) EM-Ethics tend to give

into (resist) the personal incentive by booking higher

(lower) expense accruals; and (b) in the absence of an

incentive conflict, CFOs with low (high) EM-Ethics tend to

give into (resist) the corporate incentive by booking lower

(higher) expense accruals. We also find support for a

mediated-moderation model in which CFOs? level of EM-

Ethics influences their moral disengagement tendencies

which, in turn, differentially affect their discretionary

accruals, depending on the presence or absence of incentive

conflict. Theoretical and practical implications of these

findings are discussed.

Keywords Dispositional ethics ? Earnings management ?
Incentives ? Moral disengagement

Introduction

Earnings management involves the manipulation of reve-

nues and/or expenses to obtain a desired financial reporting

outcome (e.g., Ball 2006; Healy and Whalen 1999;

Schipper 1989). This practice has played a role in the

downfall of some major corporations (e.g., Enron and

Sunbeam) and led to a push by the accounting profession

and standard setters for regulatory changes (Elias 2002;

Lawton 2007; SEC 2008). For example, in his 2002 testi-

mony before the UK Parliament Select Committee on

Treasury, International Accounting Standards Board

(IASB) Chair Sir David Tweedie decried the widespread

use of aggressive earnings management (Tweedie 2002).

Similarly in 1998, then Chair of the US Securities and

Exchange Commission (SEC), Arthur Levitt, warned that

earnings management erodes investor confidence and

undermines credibility of the financial markets (Levitt

1998), a view that is also reflected more recently by the

SEC (SEC 2008). However, despite regulatory efforts to

Electronic supplementary material The online version of this
article (doi:10.1007/s10551-014-2107-x) contains supplementary
material, which is available to authorized users.

C. A. Beaudoin

Accounting Faculty, School of Business Administration,

University of Vermont, Burlington, VT 05405, USA

e-mail: [email?protected]

A. M. Cianci (&)
Accounting Faculty, School of Business, Wake Forest

University, Winston Salem, NC 27109, USA

e-mail: [email?protected]

G. T. Tsakumis

Department of Accounting & MIS, Alfred Lerner College of

Business and Economics, University of Delaware, Newark,

DE 19716, USA

e-mail: [email?protected]

123

J Bus Ethics (2015) 128:505?518

DOI 10.1007/s10551-014-2107-x

combat aggressive financial reporting (e.g., Sarbanes?Ox-

ley Act of 2002), earnings management persists and is

exacerbated by managers? incentives (e.g., Cohen et al.

2008; McVay 2006). Thus, it is important to understand

earnings management and investigate ways to minimize its

potentially dysfunctional effects (SEC 2008).

To investigate these issues, we conduct an experiment to

examine the joint effect of incentive conflict (i.e., the

presence or absence of a personal financial incentive that

conflicts with a corporate financial incentive) and chief

financial officers? (hereafter ??CFOs??) assessments of the

ethicalness of key earnings management motivations

(hereafter ??EM-Ethics,?? dichotomized as high or low) on

earnings management behavior. In our setting, a personal

financial incentive is an incentive to increase current period

expenses to maximize bonus potential over a two-year

period and a corporate financial incentive is an incentive to

minimize expenses to achieve corporate targets. We

manipulate incentive conflict, because prior research has

found that incentives play an important role in earnings

management behavior (Bergstresser and Philippon 2006;

Burns and Kedia 2006; Ibrahim and Lloyd 2011).
1
Our

measure of EM-Ethics, developed specifically for this

study, is a fourteen-item construct based on executives?

motivations for managing earnings identified in the seminal

survey conducted by Graham et al. (2005). We focus on

EM-Ethics, a dispositional measure, because, as suggested

by Al-Khatib et al. (2004), the individual is the correct unit

of analysis when investigating ethics since it is the indi-

vidual?s ??personal?? code of ethics that ultimately influ-

ences his/her behavior. This notion is especially relevant to

the current context given the varying perspectives on

earnings management, with some viewing it as an unethical

practice resulting in negative consequences (e.g., Johnson

et al. 2012; Kaplan 2001; Vinciguerra and O?Reilly-Allen

2004), while others suggesting that it is an inherent result

of the financial reporting process that does not eliminate

the usefulness of accounting earnings (e.g., Graham et al.

2005; Lin et al. 2012; Parfet 2000). Further, we examine

CFOs? assessment of EM-Ethics in particular because the

CFO is the company?s financial reporting gatekeeper,

responsible for approving actions that may lead to earnings

management (Levitt 2003) and contributing, along with

other executives, to creating a ?tone at the top? that shapes

the ethical culture and climate within the organization (e.g.,

Sweeney et al. 2010; Arel et al. 2012).

Prior research finds that CFOs make accrual decisions

consistent with maximizing their personal incentives (e.g.,

Cohen et al. 2008; Fields et al. 2001). Our results only

provide directional (not statistically significant) support for

the expectation that in the presence (absence) of a personal

financial incentive that conflicts with a corporate financial

incentive, CFOs tend to engage in more (less) self-inter-

ested earnings management. However, consistent with our

hypotheses, we find that CFOs? EM-Ethics moderates their

willingness to manage earnings under either incentive

conflict condition. Specifically, we find that (a) in the

presence of a personal financial incentive that conflicts

with a corporate financial incentive, CFOs with low (high)

EM-Ethics tend to give into (resist) the personal incentive

by booking higher (lower) expense accruals; and (b) in the

absence of a personal financial incentive that conflicts with

a corporate financial incentive, CFOs with low (high) EM-

Ethics tend to give into (resist) the corporate incentive by

booking lower (higher) expense accruals. Also consistent

with our hypotheses, we find support for a mediated-

moderation effect whereby CFOs? EM-Ethics significantly

influences their propensity to morally disengage morality

from their actions and give into incentives. That is, the

propensity to morally disengage differentially affects the

level of CFOs? expense accruals depending on their

incentives. CFOs with high (low) EM-Ethics are less

(more) likely to morally disengage and thus give into a

personal financial incentive (i.e., book larger expense

accruals) or a corporate financial incentive (i.e., book

smaller expense accruals).

Our findings contribute to the literature in several ways.

First, we provide the first experimental evidence of the

joint impact of incentives and dispositional EM-Ethics on

CFOs? earnings management decisions. While prior

research has examined incentive contract effects (e.g.,

Ghosh and Olsen 2009; Healy 1985; Holthausen et al.

1995), no prior studies, to our knowledge, have examined

CFOs? incentives in conjunction with an individual dif-

ference variable such as EM-Ethics. Our results suggest

that the EM-Ethics/earnings management relation is mod-

erated by the presence or absence of incentive conflict.

Second, we develop a dispositional measure (EM-Ethics)

and provide experimental evidence of its impact on CFOs?

earnings management behavior. In this way, we extend

prior survey research on attitude differences related to the

ethical acceptability of earnings management among un-

dergrads, MBAs, and practicing accountants (e.g., Fisher

and Rosenzweig 1995; Greenfield et al. 2008; Kaplan et al.

2012). Prior studies examining the link between general

individual differences and ethical decision making in

business settings provide mixed results (e.g., Carpenter and

Reimers 2005; Maroney and McDevitt 2008; Mintchik and

Farmer 2009). This stream of literature has not provided

evidence that context-specific individual differences are

linked to context-specific behavior, which may be an

1
Additionally, extensive prior research on agency theory provides

evidence of the conflicting incentives present in the principal?agent

relation (e.g., Ettredge et al. 2013; Fischer and Louis 2008; Pierce

2012).

506 C. A. Beaudoin et al.

123

explanatory factor of why mixed results have been found in

business settings. The current study provides insight into

this issue by demonstrating that EM-Ethics, a context-

specific individual difference variable, affects CFOs?

earnings management behavior. Finally, we also provide

evidence of how CFOs? EM-Ethics operates through their

propensity to morally disengage. This is consistent with

recent research highlighting moral disengagement as an

individual cognitive orientation that significantly affects

unethical behavior (Moore et al. 2012). Specifically, CFOs?

level of EM-Ethics influences their moral disengagement

tendencies which, in turn, differentially affect their dis-

cretionary accruals, depending on the presence or absence

of incentive conflict.

The rest of the paper is organized as follows. The next

section reviews relevant literature and presents our

hypotheses. In the subsequent sections, we describe our

research method and present our results. We conclude with

a discussion of the implications and limitations of our

research and offer suggestions for future research.

Literature Analysis and Hypotheses Development

Earnings Management and Incentives

Earnings management is one example of an agency cost

where the misalignment of interests between the agent

(e.g., manager) and principal (e.g., firm, superior, and

shareholders) leads the agent to maximize his/her own

economic interests at the expense of the principal (Eisen-

hardt 1989; Jensen and Meckling 1976). One way to

manage earnings is to manipulate revenues or expenses by

making income-increasing or income-decreasing discre-

tionary accruals (Levitt 1998; Noronha et al. 2008). Man-

agers? discretionary accruals tend to be income decreasing

when managers have incentives to defer earnings and

income increasing when managers have incentives to

accelerate earnings. Using expenses as an example, man-

agement may overestimate costs when the company is

profitable and exceeds its financial targets or underestimate

costs to maximize earnings in the current period. These

actions may be undertaken to avoid falling short of a bonus

threshold or earnings target or to improve the issue price

around an IPO (Chung et al. 2005; Cohen et al. 2008;

Guidry et al. 1999; Healy 1985; Holthausen et al. 1995;

Matsunaga and Park 2001; Shaw 2003; Teoh et al. 1998).

In an effort to minimize earnings management, organi-

zations may focus on the structure of compensation con-

tracts, which frequently include a base salary plus a cash

bonus (Crocker and Slemrod 2007; Evans and Sridhar

1996; Jensen and Meckling 1976; Watts and Zimmerman

1986). A cash bonus can be either fixed as a percentage of

salary (i.e., a retention bonus) or variable (i.e., based on the

agent achieving certain financial targets).
2
In our experi-

mental setting, it is the variable bonus aspect of the com-

pensation contract that provides a personal incentive to

manage earnings via self-interested discretionary accruals.

For example, if prior to making an expense accrual deci-

sion, a CFO knows that projected expenses for the current

year are favorable relative to his/her variable bonus targets

(i.e., below bonus targets), s/he may record additional

discretionary accruals to reduce incurred expenses in the

subsequent year, thereby gaming the system to maximize

his/her combined two-year bonus payout. Conversely,

when bonus targets are guaranteed as a fixed percentage of

salary (i.e., personal financial incentives are absent), the

CFO?s financial reporting decisions are influenced pri-

marily by the corporate financial targets set by the execu-

tive management group (i.e., corporate financial

incentives). Thus, the use of accounting discretion to

maximize either personal or corporate financial incentives

is considered earnings management.

In our setting, larger discretionary expense accruals will

maximize the potential bonus payout over a 2-year period

but will conflict with corporate financial incentives to

minimize overall expenses. Thus, we expect that when

CFOs have a personal financial incentive that conflicts with

a corporate financial incentive, they will book larger dis-

cretionary expense accruals than when they don?t have

conflicting incentives. That is, we expect CFOs to book

larger (smaller) discretionary expense accruals [represen-

tative of income-decreasing (increasing) earnings man-

agement] when a personal financial incentive that conflicts

with a corporate financial incentive is present (absent). This

leads to the following hypothesis:

H1 In the presence (absence) of a personal financial

incentive that conflicts with a corporate financial incentive,

CFOs will record larger (smaller) discretionary expense

accruals.

The Interaction of Incentive Conflict and EM-Ethics

Prior earnings management research has primarily focused

on earnings management in a capital markets setting,

examining the influence of institutional and other factors

on its practice, detection, magnitude, and consequences

(e.g., Bedard et al. 2004; Habib and Hansen 2008; Fan

et al. 2010; Krishnan 2003; Lee 2012; Xiong et al. 2010).

Some survey and experimental research have investigated

2
Fixed bonuses are often called ??retention?? or ??stay?? bonuses which

are used as an incentive to retain key employees (e.g., Phadnis 2013;

Scholtes 2009; Smith and Pleven 2009; Lublin 2013). Such bonuses

have become increasingly popular (Klaff 2003) with, for example,

Yahoo and Starbuck CEOs receiving millions of dollars of such

bonuses in recent years (Isidore 2013; Smith 2012).

The Impact of CFOs? Incentives and Earnings Management Ethics 507

123

earnings management-related attitudes and ethical percep-

tions of academics, accountants, and students (e.g., Elias

2002; Fisher and Rosenzweig 1995; Greenfield et al. 2008;

Kaplan 2001; Kaplan et al. 2012). However, prior research

has not examined whether ethical assessments of earnings

management are associated with accountants? earnings

management behavior. In the current study, we address this

gap in the literature by examining the ethical perceptions of

earnings management (i.e., EM-Ethics) and their interac-

tive influence with incentives on CFOs? earnings man-

agement behavior.

Perceptions of the ethicalness of earnings management

vary. On the one hand, some view earnings management as

an unethical practice resulting in negative consequences.

For example, some contend that earnings management,

??probably the most important ethical issue facing the

accounting profession?? (Merchant and Rockness 1994,

p. 92), obscures true firm value and erodes trust between

shareholders and companies (e.g., Graham et al. 2006;

Levitt 1998; Loomis 1999; Huang et al. 2008). On the other

hand, others suggest that earnings management is a nec-

essary and logical result of the flexibility in financial

reporting options, with managers routinely choosing

among all options permissible under GAAP in an effort to

maximize shareholder value (e.g., Parfet 2000; Chambers

and Lacey 1996; Dobson 1999; Dye 1988; Schipper 1989).

Further, while many reasons for earning management

behavior have been identified by prior research (Graham

et al. 2005), assessments of the overall ethicalness of

earnings management may vary depending on perceptions

underlying its purpose. For instance, prior research sug-

gests that managing earnings for self-interested purposes

are perceived as less ethical than managing earnings for the

benefit of the company (e.g., Kaplan 2001; Merchant and

Rockness 1994). Thus, individuals may differ in their

perceptions of the ethicalness of earnings management.

According to ethical decision-making models, ethical

perceptions are influenced by one?s ethical sensitivity

and the context of the judgment/issue and, in turn, these

perceptions influence ethical behavior (Jones 1991; Rest

1979; Trevin~o 1986). Consistent with this notion,

accounting research finds that higher levels of ethical

reasoning and moral intensity and greater sensitivity to

shareholders? interest are negatively associated with

aggressive accounting decisions (e.g., Arel et al. 2012;

Maroney and McDevitt 2008; Ponemon 1992). Applying

this research to the current context, we suggest that

CFOs? different perceptions of the ethicalness of key

earnings management motivations?i.e., their EM-Eth-

ics?will affect their propensity to engage in this prac-

tice when they are presented with incentives to do so.

While prior research suggests that both incentives and

ethical assessments influence earnings management (e.g.,

Chung et al. 2005; Greenfield et al. 2008; Guidry et al.

1999; Healy 1985; Kaplan 2001; Kaplan et al. 2012), no

research, to our knowledge, has examined the joint effect

of these two variables on earnings management behavior.

Specifically, prior research indicates that managers use

accounting discretion to manage earnings in order to

maximize cash bonuses (e.g., Guidry et al. 1999; Healy

1985; Ibrahim and Lloyd 2011) and equity compensation

(e.g., Bergstresser and Philippon 2006; Burns and Kedia

2006; Cheng and Warfield 2005). In addition, prior

research documents that the perception of the ethicalness of

a given issue influences accounting decisions (e.g., Arel

et al. 2012; Maroney and McDevitt 2008). However, in the

current study, we examine the perceived ethicalness of key

earnings management motivations?EM-Ethics?and, to

our knowledge, no research has examined the joint effect

of incentives and EM-Ethics on earnings management

behavior.

Per Hypothesis 1, we expect CFOs with a personal

financial incentive that conflicts with a corporate finan-

cial incentive to record larger discretionary expense

accruals than CFOs without such a conflicting incentive.

Further, recall that in our setting, corporate incentives are

to minimize expenses to help the company meet corpo-

rate financial targets, while personal incentives are to

shift future period?s expenses into the current year in an

effort to maximize their bonus potential over a two-year

period. Therefore, we posit that when presented with a

personal financial incentive that conflicts with a corporate

financial incentive, low (high) EM-Ethics CFOs will

record larger (smaller) expense accruals, representative

of more (less) self-interested earnings management.

When not presented with a conflicting personal financial

incentive, CFOs with low (high) EM-Ethics will record

smaller (larger) expense accruals, representative of more

(less) company-related earnings management. Thus, we

expect that high EM-Ethics CFOs will resist giving into

either a personal or corporate financial incentive. In

contrast, we expect that low EM-Ethics CFOs will give

into these incentives and manage earnings accordingly.

Based on this discussion, we hypothesize the following

interaction:

H2 EM-Ethics and incentive conflict will interact such

that in the presence (absence) of a personal financial

incentive that conflicts with a corporate financial incentive,

CFOs with low EM-Ethics will record larger (smaller)

discretionary expense accruals as compared to CFOs with

high EM-Ethics.

508 C. A. Beaudoin et al.

123

The Mediating Role of Moral Disengagement

We posit that moral disengagement is the mechanism

through which the interaction between EM-Ethics and

incentives is activated, with high (low) moral disengage-

ment propensity exacerbating (diminishing) earnings

management behavior. That is, we expect that CFOs? EM-

Ethics will significantly influence their tendencies to mor-

ally disengage and give into incentives. Moral disengage-

ment propensity will, in turn, differentially affect the level

of CFOs? expense accruals depending on the presence or

absence of incentive conflict. Moral disengagement occurs

through a set of eight interrelated cognitive mechanisms

that allow an individual to disengage self-sanctions that

govern his/her behavior (Bandura 1986, 1991, 2002).
3

According to Bandura (1999), people adopt moral stan-

dards (e.g., ideals and values) which, when activated, serve

as self-reactive deterrents for unethical behavior. However,

individuals use strategies to rationalize, justify, or down-

play their unethical choices?i.e., to disengage their moral

standards from their conduct?thereby protecting their

self-image, minimizing cognitive distress and allowing

them to act unethically (Bandura et al. 1996). Moral dis-

engagement theory has been used to explain why individ-

uals knowingly engage in socially inappropriate/delinquent

behaviors (e.g., Moore et al. 2012; Naquin et al. 2010) and

what cognitions underlie various self-serving behavior such

as corporate wrong doing, corruption, and political vio-

lence (e.g., Bandura 1990; Moore 2008).

We suggest that an individual?s EM-Ethics is inversely

related to their propensity to disengage their personal moral

standards from their conduct. For example, CFOs with high

EM-Ethics should be generally less willing to view earn-

ings management as an acceptable practice. As a result,

high EM-Ethics CFOs will be more likely to activate their

own personal moral standards, making it more difficult for

them to adopt strategies to rationalize/downplay unethical

behavior; in this way, tendencies to morally disengage or

deactivate their personal moral standards will be reduced

when faced with incentives to manage earnings. Con-

versely, CFOs with low EM-Ethics should be generally

more willing to view earnings management in a favorable

light. As a result, low EM-Ethics CFOs will be less likely

to activate their own personal moral standards, making it

easier for them to adopt strategies to rationalize/downplay

unethical behavior; in this way, their personal moral stan-

dards and any self-sanctions related to engaging in uneth-

ical behavior (including aggressive earnings management)

will be disengaged.

Thus, we expect CFOs with high (low) EM-Ethics to

exhibit lower (higher) tendencies to morally disengage and

give into incentive-consistent behavior. In turn, CFOs with

lower moral disengagement tendencies will make smaller

(larger) expense accruals in the presence (absence) of a

personal financial incentive that conflicts with a corporate

financial incentive, thereby overriding both personal and

company incentives to manage earnings. CFOs with higher

moral disengagement tendencies will make larger (smaller)

expense accruals in the presence (absence) of a personal

financial incentive that conflicts with a corporate financial

incentive, thereby pursuing the achievement of both per-

sonal and company incentives to manage earnings. This

proposed model is presented in Fig. 1.
4
This model sug-

gests that CFOs? EM-Ethics influences their propensity to

morally disengage, which in turn differentially affects

CFOs? expense accruals depending on the presence or

absence of incentive conflict. We, therefore, hypothesize

the following effect:

ACCRUAL

INCENTIVE
CONFLICT

MORAL
DISENGAGEMENT

EM-ETHICS

Fig. 1 Mediated-moderation model

3
Bandura (1999) posits that the following eight cognitive mecha-

nisms facilitate unethical behavior: moral justification (reframing

unethical acts as being in support of the greater good?e.g., redefining

the morality of killing to justify military action), euphemistic labeling

(using sanitized language to rename harmful actions and make them

appear more benign?e.g., fired employees described as being given a

??career alternative enhancement??), advantageous comparison (con-

trasting the behavior under examination with more reprehensible

behavior to make the former seem innocuous?e.g., ??The Vietnam

war saved the populace from communist enslavement??), displace-

ment of responsibility (attribution of personal responsibility to

authority figure[s]?e.g., Nazi prison guards claiming they were just

carrying out orders), diffusion of responsibility (attribution of personal

responsibility across members of a group?e.g., requiring a group

decision to get otherwise considerate people to behave unethically),

distortion of consequences (minimizing the seriousness of the effects

on one?s actions?e.g., moving a person far away from destructive

results to weaken the potential injurious effects on that person),

dehumanization (framing the victims of one?s actions as undeserving

of basic human consideration?e.g., during wartime, nations casting

their enemies as ??demons?? or ??beasts??), and attribution of blame

(assigning responsibility to the victims themselves?e.g., computer

hackers explaining that they are forced to hack into government

databases because of a villainous government).

4
The results and variables presented in Fig. 1 are discussed in the

??Results?? section.

The Impact of CFOs? Incentives and Earnings Management Ethics 509

123

H3 Incentive conflict will moderate the relationship

between CFOs? moral disengagement tendencies and their

discretionary expense accrual decisions, such that CFOs?

moral disengagement tendencies are influenced by their

individual EM-Ethics levels.

Research Method

Participants

Participants are 83 experienced financial statement preparers

(i.e.,financialofficers with the title ofCFOorequivalent) with

an average of 28.53 years of professional work experience.
5
It

was important that we select experienced executives (such as

these) who play a key role in the financial reporting decisions

of their companies since our experiment asks participants to

assume the role of a company controller faced with a discre-

tionary expense accrual decision. Of the 83 participants, 65 %

have current or prior experience working at publicly traded

companies, while 74 %