Determinants of audit report lag: A meta‐analysis

• DOI: http://doi.org/10.1111/ijau.12136
• Author: Ahsan Habib,Muhammad Shahin Miah,Hedy Jiaying Huang,Md. Borhan Uddin Bhuiyan
• Published date: 01 March 2019
• Date: 01 March 2019

ORIGINAL ARTICLE

Determinants of audit report lag: A meta‐analysis

Ahsan Habib |Md. Borhan Uddin Bhuiyan |Hedy Jiaying Huang |

Muhammad Shahin Miah

School of Accountancy, Massey University,

Auckland, New Zealand

Correspondence

Ahsan Habib, School of Accountancy, Massey

University, Auckland, New Zealand.

Email: a.habib@massey.ac.nz

This paper provides a meta‐analysis of the determinants of audit report lag, defined as

the period between a company's fiscal year end and the audit report date. We group

the meta‐analyzed studies into three categories: (a) audit and audit‐related determi-

nants, (b) corporate governance‐related determinants, and (c) firm‐specific determi-

nants. We find that audit opinion and audit season variables increase audit report

lag, whereas Big 4 affiliation, nonaudit services, and auditor tenure decrease audit

report lag. Among the corporate governance determinants, the existence of a financial

expert member on an audit committee, and ownership concentration, reduce audit

report lag. Finally, an examination of firm‐level characteristics reveals that firm com-

plexity increases audit report lag, whereas profitability reduces it. We employ a

meta‐regression technique and identify publication bias. Although we find some evi-

dence of journal quality as a contributor to publication bias, the extent of publication

bias from this source is small.

KEYWORDS

audit report lag, corporate governance, external audit

1|INTRODUCTION

The purpose of this paper is to provide a meta‐analytic review of the

determinants of audit report lag (ARL). In their seminal article, M. C.

Jensen and Meckling (1976) argued that a well‐structured corporate

governance system is essential for mitigating agency costs emanating

from the divergence of interest between professional managers and

shareholders. One purpose of a governance structure is to ensure

the credibility of externally reported financial statements, a topic of

extensive research (Cohen, Krishnamoorthy, & Wright, 2004). The

external audit serves as a monitoring device and is, thus, a crucial

part of the corporate governance structure. Auditors are considered

to be watchdogs, since an external auditor can build a reputation

only by providing an independent verification of the financial state-

ments prepared by corporate management (Watts & Zimmerman,

1983, 1986). External stakeholders consider audit reports to be of

great value, and hence the timing of the release of an audit report,

proxied by the ARL, becomes an important input for investment

decision‐making.

ARL is defined as the period between a company's fiscal year end

and the audit report date, and it is one of the few externally

observable audit output variables that allow outsiders to gauge audit

efficiency (Bamber, Bamber, & Schoderbek, 1993). Because the audit

report contains the auditor's opinion regarding the credibility of the

financial statements, investors generally prefer short ARL. In the

USA, the first regulation stipulating a 90‐day time period after the fis-

cal year end for submitting the audited annual reports came in 1970

(Bryant‐Kutcher, Peng, & Weber, 2013). In September 2002, the Secu-

rities & Exchange Commission (SEC) adopted a new regulation reduc-

ing the filing deadline to 75 days for registrants meeting the

“accelerated filers”criteria.

1

The SEC argued that advancement of

information technology and accounting systems should enable firms

to file more quickly, an act that would facilitate more efficient valua-

tion and pricing of securities (SEC, 2002). However, Bryant‐Kutcher

et al. (2013) found an increase in subsequent accounting restatements

for firms that are required to file more quickly. Lambert, Jones, and

Brazel (2011) found that the acceleration of filing deadlines increased

discretionary accruals, suggesting that the quick filing deadline may

have impacted the auditors' ability to detect material misstatements.

Further regulatory changes regarding the filing deadline took effect

for the fiscal year beginning December 15, 2006, when the SEC fur-

ther accelerated the 10‐K filing deadline by decreasing the statutory

Received: 19 August 2016 Revised: 17 May 2018 Accepted: 24 June 2018

DOI: 10.1111/ijau.12136

20 © 2018 John Wiley & Sons Ltd Int J Audit. 2019;23:20–44.wileyonlinelibrary.com/journal/ijau

due date from 75 to 60 days for firms considered “large accelerated

filers.”

2

Outside the USA, the adoption of International Financial

Reporting Standards as well as the implementation of new Chinese

accounting standards have been found to affect the ARL (Habib,

2015; Habib & Bhuiyan, 2011).

Since the ARL is expected to vary cross‐sectionally, because of

firm and audit‐specific characteristics, an understanding of the possi-

ble determinants of ARL should provide insights into audit efficiency.

Prior research on ARL focuses on identifying and expanding the set

of variables likely to explain the ARL in the USA, as well as in countries

outside the USA. General findings from this research indicate that ARL

is affected by audit and auditor attributes (e.g., auditor affiliation, audi-

tor tenure, nonaudit services [NASs], going‐concern opinion, and audi-

tor changes), firm‐specific fundamental variables (e.g., the complexity

of the audit due to client size, foreign operations, or number of subsid-

iaries), client financial condition (existence of loss and/or distress risk),

and organizational risk (e.g., leverage). However, there remains large

variation in the reported results with respect to the determinants of

ARL across and within countries.

A recent paper by Abernathy, Barnes, Stefaniak, and Weisbarth

(2017) has provided a first comprehensive review on the determinants

of ARL in an international setting. Abernathy et al. (2017) concluded

from their review that “the ARL literature as it relates to audit charac-

teristics provides evidence that companies audited by Big Nauditors

and industry specialist auditors have shorter ARL.”Yet our meta‐analy-

sis reports that about 53% of the published results provide insignificant

coefficients on the Big 4 variable, despite the compelling theoretical

arguments that Big 4 auditing reduces ARL. The same holds for the

“busy season”variable, where 54% of the studies report insignificant

coefficients whilst 16% of the studies report a coefficient that is con-

trary to the expected positive sign. Such mixed results are also reported

for other audit, corporate governance, and firm‐specific determinants

of ARL (to be explained in Section 4). Unlike literature reviews on ARL

that summarize prior research, irrespective of the quality of its publica-

tion outlets, meta‐analysis is often used to summarize prior research

findings that seem mixed. It seems appropriate to apply meta‐analysis

to the ARL setting, to reconcile the conflicting findings across countries

and jurisdictions (Hay, Knechel, & Wong, 2006; Hwang & Lin, 1999;

Khlif & Chalmers, 2015; Lin & Hwang, 2000). In particular, we test for

the presence or absence of publication bias using a meta‐regression

analysis technique (Hay & Knechel, 2017).

Although narrative literature reviews may include a large number

of studies on particular research themes, such reviews can be mislead-

ing and often inconclusive (Hunter & Schmidt, 1990; Lin & Hwang,

2000). In some cases, there may be several studies with varying results

that are subject to variations in sample size, time period, and setting.

As a result, different researchers may reach different conclusions

about a set of individual studies. By contrast, meta‐analysis aggregates

results statistically across individual studies and corrects for statistical

artefacts like sampling and measurement errors, thereby providing

much greater precision with respect to the findings, compared with

narrative reviews (Hay et al., 2006; Lin & Hwang, 2000).

Despite the importance of conducting meta‐analysis, there has

been little such research in the accounting discipline, and particularly

in the subdiscipline of auditing. Khlif and Chalmers (2015) identified

27 previous meta‐analyses in the accounting discipline, with only

seven in the area of auditing. Prior meta‐analyses in auditing include

Trotman and Wood (1991) on internal control evaluation judgements

by different auditors, Kinney and Martin (1994) on audit‐related

adjustments and pre‐audit earnings and assets, Hay et al. (2006) and

Hay (2013) on client, auditor, and engagement attributes and audit

fees, Lin and Hwang (2010) on audit quality and earnings manage-

ment, Habib (2012) on nonaudit fees (NAFs) and accounting informa-

tion quality, and Habib (2013) on the determinants of audit opinion

decisions.

We employ the Stouffer combined test to identify the potential

determinants of ARL. We aggregate results statistically across 59 pub-

lished studies and a sufficiently large number of explanatory variables.

We categorize the potential determinants of ARL into (a) auditor and/

or audit engagement characteristics, (b) corporate governance charac-

teristics, and (c) firm‐specific characteristics. Meta‐analysis results indi-

cate that, among the audit‐related variables, audit fees, audit opinion,

auditor change, auditing season, and internal control weakness (ICW)

increase the ARL, whereas auditor‐provided NASs decrease the ARL.

With respect to corporate governance characteristics, we find that

ownership concentration and board independence reduce ARL,

whereas chief executive officer (CEO) duality increases ARL. Finally,

firm‐level characteristics across a number of studies provide generally

consistent evidence that firm complexity increases ARL and profitabil-

ity reduces it. Importantly, ARL increases for firms reporting losses,

extraordinary items, and announcing accounting restatements. A novel

contribution of our meta‐analysis is the application of a meta‐regres-

sion technique recently applied in meta‐analysis on the magnitude of

the fee premium for Big 4 auditors (Hay & Knechel, 2017). We find

the presence of publication bias in our meta‐analyzed results. We then

break down the bias into some contextual settings. Although we find

some evidence of journal quality as a contributor to publication bias,

the extent of publication bias from this source is small.

We contribute to the auditing literature by applying a meta‐anal-

ysis technique to an important audit output variable, ARL. Many of

these findings have policy implications. For example, restrictions on

NASs have been imposed in order to protect auditor independence,

yet our meta‐analysis shows that high levels of NASs actually lead to

a shorter ARL, and this often implies lower audit fees and less prob-

lematic audits (Knechel & Payne, 2001; Knechel & Sharma, 2012). This

is evident in studies within the USA, where the SEC has already

prohibited most NASs offered by the incumbent auditors, and in stud-

ies outside the USA. Globally, audit regulation is encouraging a change

of auditor at regular intervals to ensure the release of timely audit

reports. However, our findings suggest that changing auditors has an

adverse effect on ARL, implying that it takes time for a newly

appointed auditor to become familiarized with the client.

Our findings also reveal that a number of corporate governance

components (i.e., financial expertise on the audit committee, board

independence, and CEO duality) are more effective in improving audit

reporting timelines. In July 2015, the SEC called for input on “possible

revisions to audit committee disclosures”and for more research in

understanding the attributes of the audit committee members that

contribute to effective oversight of the financial reporting process in

listed companies (SEC, 2015). Apparently, the US regulators still have

HABIB ET AL.21