An Investigation of Auditors’ Judgments When Companies Release Earnings Before Audit Completion

• Published date: 01 May 2019
• Author: JOSEPH H. SCHROEDER,LORI SHEFCHIK BHASKAR,PATRICK E. HOPKINS
• Date: 01 May 2019
• DOI: http://doi.org/10.1111/1475-679X.12262

DOI: 10.1111/1475-679X.12262

Journal of Accounting Research

Vol. 57 No. 2 May 2019

Printed in U.S.A.

An Investigation of Auditors’

Judgments When Companies

Release Earnings Before Audit

Completion

LORI SHEFCHIK BHASKAR,

∗PATRICK E. HOPKINS ,∗

AND JOSEPH H. SCHROEDER

∗

Received 1 December 2017; accepted 30 December 2018

ABSTRACT

The majority of U.S. public companies release annual earnings prior to

the completion of audit fieldwork. We investigate this phenomenon in a

controlled experiment with audit partners and senior managers. We find

that releasing earnings before completion of the audit pressures auditors

to adopt the goals of management, thereby reducing the likelihood of post-

announcement audit-adjustment recommendations. We also examine the ef-

fect of audit committee (AC) strength in improving auditors’ judgments after

∗Indiana University.

Accepted by Christian Leuz. We thank Mary Barth, Dirk Black, Kathryn Kadous, Justin

Leiby,Mark Peecher, Kathy Rupar, Devin Williams, Donnie Young,and an anonymous referee;

workshop participants at Ball State University, Georgia Institute of Technology, University of

Florida, University of Mannheim, Monash University, and Penn State University; and partic-

ipants at the 2018 American Accounting Association Auditing Midyear Meeting (Portland),

the 2018 European Accounting Association Annual Congress (Milan), the 2017 IAAER-CIMA

Paper Development Workshop, the 2018 International Symposium on Audit Research (Maas-

tricht), the 2018 Journal of Accounting Research Conference, and the 2018 University of Utah

Winter Accounting Conference for their helpful suggestions. We also thank professionals at

EY and Michelle Hutchens for their assistance in developing the case materials, and the Cen-

ter for Audit Quality and the Kelley School of Business at Indiana University for their generous

support for this research project. Professor Hopkins thanks SungKyunKwan University for fi-

nancial support.

355

CUniversity of Chicago on behalf of the Accounting Research Center,2019

356 L.S.BHASKAR,P.E.HOPKINS,AND J.H.SCHROEDER

annual earnings are released. When ACs are actively involved in accounting is-

sues and proactively communicating with auditors—characteristics currently

lacking in most ACs—the negative effects on auditors’ judgments are com-

pletely mitigated. Our study provides evidence on potential unintended con-

sequences of early release of earnings and the importance of investing in

high-quality ACs to mitigate adverse effects of client pressures on audit judg-

ment and financial reporting quality.

JEL codes: D80; D91; G10; G30; M40; M41; M42

Keywords: auditor judgment; audit committees; audit completeness; direc-

tional goals; earnings announcements; financial reporting quality

1. Introduction

Using a controlled experiment, we investigate whether the audit-judgment

quality of highly experienced audit partners and senior managers is af-

fected by the differential timing of firms’ annual earnings releases and

firms’ financial statement audit completion. Our study is motivated by the

dramatic shift in firms releasing unaudited versus audited annual earnings

announcements. Prior to 2004, approximately 75% of annual earnings an-

nouncements were released on or after the audit report date (Bamber, Bam-

ber, and Schoderbek [1993], Schwartz and Soo [1996]), whereas, today, ap-

proximately 70% of U.S. public companies release earnings approximately

16 days, on average, prior to the audit report date (Schroeder [2016], Mar-

shall, Schroeder, and Yohn [2018]).1

Although recent archival studies document lower financial reporting

quality for firms releasing annual earnings prior to the audit report date

(Bronson, Masli, and Schroeder [2018], Marshall, Schroeder, and Yohn

[2018]), factors potentially contributing to the diminution in financial re-

porting quality are largely unexplored. While firms’ managers are responsi-

ble for the information contained in financial reports, we propose auditors

can have a contributing role in the observed reduction in financial report-

ing quality when earnings are announced prior to audit completion. Specif-

ically, because firms experience adverse capital market reactions when

announced unaudited accounting information is subsequently revised in

audited 10-K filings (Hollie, Livnat, and Segal [2005, 2012], Bronson et al.

[2011]), we propose firms announcing earnings prior to audit completion

may unintentionally increase pressure on auditors to avoid post-earnings

announcement adjustments during the completion of the audit.

1This shift is primarily due to audits taking 16 days longer,on average, post-2004 with added

requirements of Public Company Accounting Oversight Board (PCAOB) Auditing Standards

Nos. 2 and 3 (Bronson et al. [2011]) and managers maintaining their historical earnings re-

lease dates due to market pressures for timely earnings (PCAOB [2004a, b], Krishnan and

Yang [2009]). Furthermore, this practice persists to present day with firms marking their an-

nual financial statements in the earnings announcement as “unaudited” (Marshall, Schroeder,

and Yohn [2018]).

TIMING OF EARNINGS RELEASES AND AUDITORS’JUDGMENTS 357

Our study capitalizes on a comparative advantage of experiments to com-

plement and extend evidence from the extant archival literature. For exam-

ple, given limitations in archival data sets, prior research could not detect

or control for (1) whether, and to what extent, audits are actually incom-

plete when earnings are announced prior to the audit report date, or (2)

the potential role of the auditor in contributing to diminished financial re-

porting quality when earnings are announced prior to the completion of

audit procedures. In our experiment, we are able to manipulate the timing

of the earnings announcement in relation to the completion of the audit,

and isolate the potential effects on experienced auditors’ judgment quality

(i.e., their decision processing and ultimately their likelihood of requiring

a year-end audit adjustment for aggressive accounting treatment). In addi-

tion, we are able to control for potential selection in firms’ decisions to issue

earnings announcements before/after audit completion and investigate

the relation of those decisions to financial reporting quality. Furthermore,

we also exploit the experimental method to investigate a means of miti-

gating the deleterious effects of announced annual earnings on auditors’

judgments. Specifically, we manipulate audit committee (AC) effectiveness

to determine whether the AC attributes proposed by the PCAOB [2012]

and Center for Audit Quality (CAQ) [2013, 2016] improve post-earnings

announcement audit judgments (and financial reporting quality). Identi-

fying this mitigating mechanism is important because regulators are un-

likely to require firms to wait for audit completion to voluntarily announce

earnings.

We build on prior archival studies by providing three types of empirical

evidence unique to our research setting. First, given the opacity of audit

processes in real-world audit settings, we collect post-experimental survey

data from our participants revealing a significant amount of audit work

(e.g., conclusions about significant accounting estimates, audit review) is

often incomplete when annual earnings are released prior to the audit re-

port date. These data support the audit-completion-timing claims made

in prior archival studies and identify pre-audit completion earnings an-

nouncements (“released earnings”) as a potentially common and salient

source of pressure to avoid subsequent year-end audit adjustments. Second,

in our controlled experiment, we test theory as to how this client pressure

influences auditor judgments; specifically, we investigate whether releasing

annual earnings prior to the completion of the audit increases auditors’

adoption of client-preferred directional goals (referred to as “directional

goals” throughout). Third, we ex ante test whether stronger ACs, which are

unavailable in archival settings, can help auditors avoid succumbing to re-

leased earnings-induced pressure (and client pressures in general).

We develop a theoretical process model predicting released earnings

prior to audit completion influences auditors’ judgments by means of

their directional goals. Specifically, we posit released earnings prior to

the completion of the audit is perceived as a source of client pressure to

avoid consequences associated with subsequent earnings-related revisions.