Conditional conservatism and persistence of accrual components

• Author: Wenjun Zhang,Yutao Li
• DOI: http://doi.org/10.1002/cjas.1305
• Date: 01 March 2015
• Published date: 01 March 2015

Conditional conservatism and persistence of

accrual components

Yutao Li*

University of Lethbridge

Wenjun Zhang

Dalhousie University

Abstract

We examine whether conditional conservatism affects the

persistence of accrual components associated with varying

degrees of reliability. Using a sample of 102,416 f‌irm-year

observations from 1968 to 2006, we found that conservatism

reduces the persistence of more reliable accruals, but not

less reliable accruals. Our evidence suggests managers’

discretion in practicing conservatism convention may have

caused the low persistence of less reliable accruals. Our

results inform capital market participants, such as f‌inancial

analysts, that managers may incorporate more bad news

into more reliable accruals than in less reliable accruals,

leading to lower persistence of more reliable accruals than

less reliable accruals. We expect that analysts can improve

forecasting eff‌iciency by understanding the differential

implications of accounting conservatism on different

accrual components. Copyright © 2015 ASAC. Published

by John Wiley & Sons, Ltd.

Keywords: accounting conservatism, C –Score, accrual

persistence, accrual reliability

Résumé

Cet article examine l’impact possible du conservatisme

conditionnel sur la longévité de la composante comptabilité

d’exercice en rapport avec des degrés variables de f‌iabilité.

À partir d’un échantillon de 102 416 observations

d’exercices d’entreprises entre 1968 et 2006, l’étude montre

que le conservatisme réduit la longévité des exercices plus

f‌iables et non celle des exercices moins f‌iables. Les données

suggérent que dans la pratique de la convention du

conservatisme, la discrétion des gestionnaires serait

responsable de la faible longévité des exercices moins

f‌iables. Les résultats font comprendre aux participants du

marché des capitaux tels que les analystes f‌inanciers que

les gestionnaires sont capables d’injecter plus de mauvaises

nouvelles dans les exercices f‌iables que dans les exercices

moins f‌iables, dans le but de réduire la persistance des

exercices plus f‌iables. La prise de conscience des implica-

tions différentielles du conservatisme comptable sur les

différentes composantes des exercices devrait amener les

analystes à améliorer la qualité de leur prévision. Copyright

© 2015 ASAC. Published by John Wiley & Sons, Ltd.

Mots-clés : conservatisme comptable, C –Score, longévité

de l’exercice, f‌iabilité de l’exercice

We examine whether conditional accounting conserva-

tism affects the persistence of accrual components.

Decomposing accruals into components with varying de-

grees of reliability, Richardson, Sloan, Soliman, and Tuna

(2005) documented that less reliable accruals (e.g., nonoper-

ating accruals) have a lower ability to predict future earnings

than do more reliable accruals (e.g., f‌inancial accruals).

Although Richardson et al. discussed the possibility that

the low persistence of less reliable accruals could result from

conservative accounting practices, there is no empirical

study investigating the effects of conservatism on the persis-

tence of accrual components. In light of recent f‌indings that

conditional conservatism reduces earnings persistence

(Bandyopadhyay, Chen, Huang, & Jha, 2010; Chen,

Folsom, Paek, & Sami, 2014), we conjecture that condi-

tional conservatism may contribute to the low persistence

of less reliable accruals. We base this on the fact that

accounting conservatism implies a timely recognition of

transitory losses, which is more so the cas e in less reliable

accruals than in more reliable accruals (Guay & Ver recchia,

2006). This inquiry allows us to explore the differential effects

of conservatism on each component of accruals individually,

as opposed to its effect on the accruals as a whole.

The authors would like to thank Professor Michel Magnan (editor), an anon-

ymous referee, Dr. Tiemei Li (a discussant at the 2011 Canadian Academic

Accounting Association Annual Conference), and participants at the 2011

Canadian Academic Accounting Association Annual Conference for their

insights and helpful comments on an earlier version of this paper. Any re-

maining errors are our own.

*Please address correspondence to: Yutao Li, Assistant Professor, Faculty

of Management, University of Lethbridge, 4401 University Drive,

Lethbridge, AB, T1K 3M4. Email: yutao.li@uleth.ca

Canadian Journal of Administrative Sciences

Revue canadienne des sciences de l’administration

32:15–29 (2015)

Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/CJAS.1305

Can J Adm Sci

32(1), 15–29 (2015)Copyright © 2015 ASAC. Published by John Wiley & Sons, Ltd. 15