Does Accrual Management Impair the Performance of Earnings‐Based Valuation Models?

• Author: Jennifer L. Kao,Yao Tian,Lucie Courteau
• DOI: http://doi.org/10.1111/jbfa.12101
• Date: 01 January 2015
• Published date: 01 January 2015

Journal of Business Finance & Accounting

Journal of Business Finance & Accounting, 42(1) & (2), 101–137, January/February 2015, 0306-686X

doi: 10.1111/jbfa.12101

Does Accrual Management Impair

the Performance of Earnings-Based

Valuation Models?

LUCIE COURTEAU,JENNIFER L. KAO AND YAO TIAN∗

Abstract: This study examines how accrual manipulations affect firm valuation in the years

surrounding the passage of the Sarbanes-Oxley Act (SOX). We compare the absolutepercentage

pricing errors of RIM and DCF valuation models for a group of US firms suspected to

have engaged in accrual manipulations to avoid a small loss or a small earnings decline vs.

‘Normal’ firms matched on industry, year and size. We find that RIM can better estimate

intrinsic value than DCF for the matched Normal firms in the pre-SOX period, but not so

for accrual manipulators, and that SOX mitigates the harmful effect of accrual manipulations,

completely eliminating the difference in RIM’s accuracy advantage over DCF between Normal

firms and accrual manipulators. As a further analysis, we redefine Suspect firms as real-activity

manipulators and find a significant across-group difference in accuracy wedge in both sample

periods, implying that SOX has prompted firms to favor real-activity manipulations over accrual

manipulations.

Keywords: accrual and real-activity manipulations, firm valuation, earnings- and non-earnings-

based valuation models, valuation errors, Sarbanes-Oxley Act

1. INTRODUCTION

Ohlson (1995) and Feltham and Ohlson (1995, 1996) provide a conceptual framework

for relating accounting earnings to firm value. Since then, several empirical studies

have shown that earnings-based valuation models can give better estimates of firm

value than non-earnings-based valuation models. While accounting earnings have

regained popularity among researchers, financial analysts and investors in recent years,

evidence suggests that earnings are often subject to managerial manipulations. Such

manipulations, driven by the pressure to meet or beat earnings expectations, are

believed to have eroded the quality of earnings and led to highly publicized corporate

∗The first author is from the Faculty of Economics and Management, Free University of Bozen-Bolzano,

Bolzano, Italy. The second author is from the Department of AOIS, University of Alberta, Edmonton,

Alberta, Canada. The third author is from the Department of Accounting and Finance, San Jose State

University, San Jose, CA, USA. We would like to thank participants at the EAA and AAA Annual Meetings

and workshop participants at Bocconi University for their helpful comments on earlier versions of the paper.

Financial support for this project is provided by the University of Alberta (SSHRC 4A and Canadian Utilities

fellowship) and the Free University of Bozen-Bolzano.

Address for correspondence: Lucie Courteau, Faculty of Economics and Management, Free University of

Bozen-Bolzano, Bolzano, Italy. e-mail: Lucie.Courteau@unibz.it

C

2015 John Wiley & Sons Ltd 101

102 COURTEAU, KAO AND TIAN

scandals. The perceived erosion of financial reporting quality in turn prompted the

US Congress to pass the Sarbanes-Oxley Act (SOX) on July 30, 2002 with the stated

objective of restoring investor confidence in financial and public reporting.

The purpose of this study is to examine empirically how the presence of ac-

crual manipulations affects firm valuation. To address this research question, we

use two samples consisting of 738 and 592 firm-year observations with complete

annual financial and stock price information and forecast data from two sub-periods

surrounding the passage of SOX, i.e., 1991–2000 (pre-SOX) and 2003–2009 (post-

SOX), respectively.1For each sample period, half of the sample firms are suspected

to have engaged in small earnings manipulations in order to avoid incurring either

a loss or an earnings decline (labeled ‘Suspect’ firms hereafter) and the other half,

matched to Suspect firms on industry, year and size, are considered to be normal

(labeled ‘Normal’ firms hereafter). Our pre-SOX analysis is intended to isolate the

effect of accrual manipulations on the relative performance of the residual income

model (RIM) and discounted cash flow model (DCF) over a period that pre-dates

major financial scandals and the ensuing legislative events. The post-SOX analysis, on

the other hand, is designed to shed light on the effectiveness of SOX in improving

earnings quality from the firm valuation perspective.

We employ a matched-pair design to first compare the difference in the absolute

percentage pricing errors for the RIM and DCF valuation models across the Normal

and Suspect groups of firms by sample period, using current stock price as the

valuation benchmark and a 2% constant growth rate to estimate terminal values.

Results indicate that RIM on average enjoys a smaller accuracy advantage over DCF

among Suspect firms than the matched Normal firms in the pre-SOX period, but

there is no difference in accuracy wedge between these two groups of firms in the

post-SOX period. For Suspect firms, while RIM does not outperform DCF in the

pre-SOX period, it displays a greater ability to estimate intrinsic value than DCF

in the post-SOX period. We then extend the analysis to a multivariate setting by

regressing RIM’s accuracy advantage over DCF, defined as the absolute percentage

pricing errors under DCF minus the corresponding figure under RIM for each firm-

year observation, on an indicator variable, SUSPECT, representing Suspect firms, and

a set of eight covariates which may also have affected the relative performance of these

two models. Consistent with the univariate results, we find that the coefficient estimate

on SUSPECT is negative and significant at the 1% level in the pre-SOX period, but it is

insignificantly different from zero in the post-SOX period. These results suggest that

accrual manipulations impaired RIM’s performance prior to the introduction of SOX

and that SOX regulations have mitigated the harmful effect of accrual manipulations

on the valuation usefulness of RIM.

Our overall conclusion about changes to the pattern of across-group differences

in RIM’s accuracy advantage over DCF surrounding SOX is robust to the selection of

research design (i.e., matched-pair vs. unmatched full sample). Untabulated results

also indicate that it extends to settings where we either re-define Suspect firms based

1 We do not include years 2001 and 2002 as part of the pre-SOX period, because a series of legislative

events leading to the eventual enactment of SOX took place during those two years. Our pre-SOX results

(untabulated) nonetheless remain substantially unchanged when we re-define the pre-SOX period to

include 2001 and 2002. We also consider an alternative definition of the post-SOX period (i.e., 2003–2006

and 2009) which excludes financial crisis years (2007 and 2008). Our post-SOX results (untabulated) are

not altered qualitatively.

C

2015 John Wiley & Sons Ltd

ACCRUAL MANAGEMENT AND EARNINGS-BASED VALUATION MODELS 103

on only one earnings threshold (i.e., earnings decline avoidance or loss avoidance),

double the post-horizon growth rate from 2% to 4%, or use ex post intrinsic value

as the model benchmark. Taken together, these results highlight the importance of

considering earnings manipulations when comparing the performance of earnings-

and non-earnings-based valuation models, especially at a time when government

regulations over financial reporting are perceived to be loose.

While we have focused on accrual manipulations in this study, they are not the only

tool that firms use to manage reported earnings. As a further analysis, we exclude

firms suspected to have undertaken real-activity manipulations from the definition

of Normal firms and find that all the results continue to hold. To offer a more

symmetric treatment, we next redefine Suspect firms as either accrual or real-activity

manipulators, but retain the above narrower definition of Normal firms. Unlike the

other cases examined in the paper, we find that at the univariate level RIM no longer

outperforms DCF among Suspect firms after the passage of SOX, when we include

real-activity manipulations in the definition of Suspect firms. This result is invariant

to an alternative definition of Suspect firms to include real-activity manipulators only,

implying a likely shift towards real-activity manipulations following SOX, as reported

byCohenetal.(2008).

We contribute to the valuation literature that compares the relative performance of

RIM and DCF valuation models by taking into account the potential effect accrual

manipulations have on model inputs. Our aim is to see whether the previously

documented superiority of RIM continues to hold when earnings are managed. By

explicitly allowing for the possibility of earnings manipulations, we arguably offer

a more accurate assessment of the estimation ability of RIM vis-`

a-vis DCF models.

We also extend the earnings management literature that examines the incentive for

and the existence of earnings management, along with its market consequences.

Our contribution lies in documenting the impact of earnings management on the

usefulness of earnings vs. cash flows for valuation purposes. Finally, we complement a

large volume of SOX literature that studies changes in earnings quality surrounding

the passage of SOX by approaching this issue from the firm valuation perspective.

Findings that RIM enjoys a similar accuracy advantage over DCF for both Normal firms

and accrual manipulators in the post-SOX period corroborate most of the empirical

evidence that SOX regulations have improved the quality of reported earnings. On

the other hand, our findings that RIM does not provide better estimation of firm value

than DCF for real-activity manipulators in the post-SOX periods lend support to prior

studies which show a shift in the method of earnings manipulations, from accruals to

real activities, following SOX.

Our study is also of practical relevance. Earnings are used extensively to estimate

firm value in practice. The majority of the 400 chief financial officers (CFOs) surveyed

by Graham et al. (2005) believe that earnings, not cash flows, are the key metric used

by outside stakeholders. Skinner and Sloan (2002) find that investors use earnings

to evaluate firm performance. However, when earnings are managed, heavy reliance

on this number in firm valuation may result in inaccurate assessment, undesirable

investment decisions and misallocation of resources. Our research intends to quantify

this effect and to raise awareness among investors and practitioners about the pitfalls

of taking managed earnings at face value and using them directly in firm valuation.

The remainder of the paper is organized as follows: Section 2 reviews the relevant

literature and develops the hypothesis for the study; Section 3 discusses the research

C

2015 John Wiley & Sons Ltd