Management Changes, Reputation, and “Big Bath”—Earnings Management

• DOI: http://doi.org/10.1111/jems.12101
• Published date: 01 September 2015
• Author: Dirk Sliwka,Petra Nieken
• Date: 01 September 2015

Management Changes, Reputation, and “Big

Bath”—Earnings Management

PETRA NIEKEN

Karlsruhe Institute of Technology

Institute of Management

Karlsruhe, Germany; and UiS Business School

Department of Economics

University of Stavanger

Stavanger, Norway

petra.nieken@kit.edu

DIRK SLIWKA

Seminar of Personnel Economics and HRM

University of Cologne, Albertus-Magnus-Platz

Cologne, Germany

dirk.sliwka@uni-koeln.de

We study the effects of managerial turnover on earnings management activities in a model in

which managers care about their external reputation. We develop an overlapping generations

model showing that both outgoing and incoming managers bias reported earnings such that

typically very low returns are reported in the first period after a manager has been replaced.

Outgoing managers shift earnings forward to their last period in office as they will not benefit

from earnings realized after that. Incoming managers can have an incentive to shift earnings to

the second period in office as reported earnings will, immediately after a management change,

only be partly attributed to their own ability. Deferred compensation can reduce incentives for

earnings management.

Key words: earnings management; management turnover; reputational concerns; career con-

cerns; image concerns

1. Introduction

Accounting data are used to create financial statements and reduce the information

asymmetry between organizations and external market participants that either have

invested or are considering investing in a company. However, these data are not always

reliable because managers typically have some discretion when reporting earnings and

might have an incentive to bias reports to influence investors’ expectations.

In this paper, we study the incentives for earnings management in a theoretical

model. Following Schipper (1989), we define earnings management as an intervention

in the external reporting process to obtain some private profit. We focus on managers’

concerns about their reputation and the consequences of managerial turnover on earn-

ings management activities. As a survey of Graham et al. (2005) reveals, more than three

We thank the coeditor, two anonymous referees, and J¨

org Budde for helpful comments and suggestions.

Financial support by the Deutsche Forschungsgemeinschaft (DFG), in particular DFG Research Unit “Design

and Behavior,” FOR1371 and SFB TR/15, is gratefully acknowledged.

C2015 Wiley Periodicals, Inc.

Journal of Economics & Management Strategy, Volume24, Number 3, Fall 2015, 501–522

502 Journal of Economics & Management Strategy

fourths of the responding executives stated career concerns and external reputation as

reasons for earnings manipulation. According to the authors, reputational concerns are

considered to be even more important than short-run compensation when it comes to

meeting an earnings benchmark.

Westudy the incentives of managers to shift earnings in an overlapping generations

model where each manager is in office for two periods. Managers careabout their salaries

but also about their reputation have a high ability. Actual earnings depend on effort

exerted by the managers as well as their ability. Reported earnings can be manipulated

to some extent by shifting gains or losses to future periods respecting a clean surplus

accounting condition. During the first period a manager is in office, actual earnings do

not only depend on his own effort and ability but to some extent also on the managerial

ability of his predecessor. We show that the external market updates its assessment of

the manager’s ability based on the reported earnings, which, in turn, creates incentives

for earnings management. Whereas a manager in his last period in office brings forward

earnings to the period in which he will leave, the incoming manager has an incentive

to shift earnings away from the first period of his tenure as these will to some extent be

attributed to the ability of his predecessor. Indeed, these two effectscan lead to the often

observed “big bath” earnings management in a new manager’s first period in office.

We also study the impact of the structure of bonus payments on earnings man-

agement. Linear, earnings-based bonus schemes cannot prevent earnings management

if bonuses are paid only when the manager is in office. The reason is that there is

an inherent tension between incentives to exert effort and incentives to reduce earn-

ings management activities. But we show that deferred compensation schemes, that is,

bonus payments after the manager has left, can reduce or even eliminate incentives

for earnings management. Hence, our model provides testable hypotheses on factors

affecting the extent of earnings management activities before and after a CEO retires.

There are several other rational expectations of models dealing with earnings man-

agement and its consequences. Dye (1988), for instance, studies earnings management

also in an overlapping generations model. In contrast to our model, several generations

of shareholders, not managers, are involved. In this setting, Dye analyzes the internal

demand for earnings management of the shareholders and the external demand for ma-

nipulation to smooth earnings. Kirschenheiter and Melumad (2002) introduce a model

that studies earnings smoothing and the “big bath” phenomena. Similar to our approach

the manager can manipulate earnings during the two periods he is in office. However,

in contrast to our model the manager wants to maximize the value of the company

and does not care about his own reputation. In the equilibrium reporting strategy, the

manager either takes a “big bath” (if actual cash flow levels are quite low) or smoothes

earnings by either under- or over-reporting actual cash flows. Hence, Kirschenheiter

and Melumad offer a possible explanation for both, earnings smoothing and taking a

“big bath,” depending on the actual cash flow. But they do not investigate the impact of

managerial turnover on earnings management, especially if managers care about their

reputation. Fischer and Verrecchia (2000) develop a model that shows how a reporting

bias can affect the informativeness of a financial report. In their setting, a manager has

the opportunity to manipulate earnings in order to achieve a higher valuation of the

firm. As the market cannot perfectly anticipate the amount of the manipulation, noise

is added to the report, which reduces the information. Furthermore, Guttmann et al.

(2006) study a rational expectation model where the manager’s compensation depends

on the stock price. Hence, the manager has an incentive to manipulate earnings, but

the investors anticipate his activities. The authors show that besides a fully revealing