Affiliation of Indonesian audit firms with Big4 and second‐tier audit firms and the cost of debt

• Date: 01 November 2019
• DOI: http://doi.org/10.1111/ijau.12167
• Published date: 01 November 2019
• Author: Philippe Van Cauwenberge,Heny Kurniawati,Heidi Vander Bauwhede

ORIGINAL ARTICLE

Affiliation of Indonesian audit firms with Big4 and second-tier

audit firms and the cost of debt

Heny Kurniawati

1,2

| Philippe Van Cauwenberge

2

| Heidi Vander Bauwhede

2

1

Accounting Department, Bina Nusantara

University, Jakarta, Indonesia

2

Department of Accounting, Corporate

Finance and Taxation, Ghent University,

Ghent, Belgium

Correspondence

Heny Kurniawati, Accounting Department,

Bina Nusantara University. Jakarta, Indonesia.

Email: heny.kurniawati@ugent.be

Funding information

Hercules Foundation, Grant/Award Number:

AUGE/11/013; Indonesian Endowment Fund

for education (LPDP Indonesia), Grant/Award

Number: PRJ-726/LPDP/2014

This paper documents that in Indonesia, where litigation risk is low and foreign audit

firms can only enter the market through affiliation with a local audit firm, the appoint-

ment of a local audit firm affiliated either with a Big4 or a second-tier audit firm

reduces the cost of debt for listed companies significantly. This finding holds

irrespective of the risk profile of the client and is independent of whether or not we

control for endogenous auditor choice. There is less conclusive evidence of a differ-

ence in the magnitude of the effect of Big4 affiliation versus affiliation with a

second-tier audit firm. Our evidence is in line with the idea that Big4 and second-tier

audit firms are perceived as applying uniform quality criteria around the world,

regardless of the local circumstances in which they operate.

KEYWORDS

audit quality, Big4, cost of debt, emerging market, Indonesia, second-tier

JEL CLASSIFICATION

G21; G32; M41; M42; O16

1|INTRODUCTION

This paper investigates whether the affiliation of local Indonesian

audit firms with a Big4 or second-tier audit firm reduces the cost of

debt for listed companies in Indonesia. Foreign audit firms are allowed

to enter the Indonesian audit market but can only do so by partnering

with a local audit firm. There are several indicators suggesting serious

quality issues within the local Indonesian audit context (Dunakhir,

2016). With few exceptions, local audit firms in Indonesia are rela-

tively small and capacity constrained. Owing to lack of adequate

resources, they face challenges in providing high-quality auditing ser-

vices (The World Bank, 2010a). Anecdotal evidence from interviews

with practicing auditors reveals high levels of compliance gaps with

respect to audit planning, documentation, related party investigation,

and fraud detection (The World Bank, 2010a).

We hypothesize that the affiliation of local audit firms with a Big4

or second-tier audit firm reduces the cost of debt for the clients. The

underlying rationale is that cooperation of a local audit firm with a

Big4 or second-tier audit firm improves the quality of the audit as per-

ceived by the creditors. We first examine the effect of affiliation of

local audit firms with a Big4 audit firm. Big4 firms are widely viewed

as producing higher quality audits than non-Big4 firms (J. R. Francis,

Maydew, & Sparks, 1999). One of the reasons is that the brand name

reputation of the Big4 audit firms provides them with an incentive to

increase their audit effort. When a local Indonesian audit firm affili-

ates with a Big4 audit firm to perform the audit of an Indonesian com-

pany, we expect that the reputational concerns of the Big4 firm will

translate into pressure on the local audit firm to increase its efforts. A

second reason why Big4 firms are viewed as producing higher audit

quality is that they have more and better resources (DeFond & Zhang,

2014). We argue that, through affiliation, local Indonesian audit firms

can expand their access to expertise and tap into the resources of the

well-equipped and more sophisticated Big4 firms. The transfer of pro-

fessionalism is even embedded in the Indonesian regulation, according

to which an international audit firm partnering with a local firm should

provide training programs, impose standards on quality control, and

perform periodic quality controls. Another reason that creditors may

perceive audit services performed by local audit firms affiliated with a

Big4 firm as being of higher quality is that there is evidence that Big4

firms use tight selection and evaluation procedures in choosing their

Received: 6 July 2017 Revised: 26 April 2019 Accepted: 8 May 2019

DOI: 10.1111/ijau.12167

Int J Audit. 2019;23:387–402. wileyonlinelibrary

.com/journal/ijau © 2019 John Wiley & Sons Ltd

387

local partners (Kartikahadi, 2010) and only team up with the better

local audit firms. Collectively, these arguments suggest that local audit

firms that are affiliated with a Big4 firm provide higher quality audits

than other local audit firms do. Lenders typically rely on the financial

statements to assess borrower quality (Kim, Song, & Tsui, 2013).

Based on the foregoing argumentation, we argue that lenders per-

ceive financial statements of borrowers that are audited by a local

Indonesian audit firm that is affiliated with a Big4 firm as being of

higher quality, hereby strengthening their beliefs regarding the accu-

racy of their assessment of the borrowers' credit risk and their per-

ceived efficiency of lender monitoring (Booth, 1992). We expect that

this will be translated in a lower cost of debt.

Next, we investigate whether affiliation of local Indonesian audit

firms with second-tier audit firms also has a negative effect on the

cost of debt. The increased demand for high-quality audits since the

Sarbanes–Oxley Act in 2002, combined with the capacity constraints

of the Big4 firms, created an increased demand for the services of

second-tier audit firms in the USA, resulting in their relatively rapid

growth (Boone, Khurana, & Raman, 2010; Jenkins & Velury, 2011).

The growth of second-tier firms and the emergence of “second tier”

as a brand name suggest that second-tier audit firms also face reputa-

tion concerns, which would reduce the differences in their incentive

system to provide high-quality audits compared with that of the Big4

audit firms. Boone et al. (2010), investigating a US sample for the

period 2003–2006, found small differences in actual audit quality

between Big4 and second-tier clients. Using the ex-ante equity risk

premium as a measure of perceived audit quality, however, they still

found a pronounced difference in favor of Big4 clients. Jenkins and

Velury (2011) found no difference in conservatism between Big4 and

second-tier clients. Cassell, Giroux, Myers, and Omer (2013) reported

that, in the post-Andersen era, the perceived financial reporting credi-

bility of second-tier clients is higher than that of other non-Big4 cli-

ents and indistinguishable from that of Big4 clients. Inspired by these

observations, we are interested to find out if affiliation of local audi-

tors with a second-tier audit firm has a negative effect on the cost of

debt and whether or not this effect is similar to the effect of affiliation

with a Big4 audit firm.

We try to answer these questions using a sample of Indonesian

listed companies over the period 2008–2015. We regress these com-

panies' costs of debt on an ordered categorical variable indicating

whether their local Indonesian audit firm was affiliated with (1) a Big4

audit firm, (2) a second-tier audit firm, or (3) none of the two, together

with various control variables, including interest coverage, profitabil-

ity, leverage, asset tangibility, size, the occurrence of negative equity,

age, growth, and cash flow performance. In line with prior studies

(J. Francis, LaFond, Olsson, & Schipper, 2005; Minnis, 2011; Vander

Bauwhede, De Meyere, & Van Cauwenberge, 2015), we proxy the

cost of debt by the effective interest cost; that is, the ratio of the

interest expense to the average amount of financial debt.

Endogeneity concerns stemming from auditor self-selection might

yield inconsistent coefficient estimates when estimating the model

using ordinary least-squares (OLS) estimation (Heckman, 1976, 1978).

Therefore, besides using OLS, we also estimate a so-called Heckman

treatment effect regression, which is a Heckman selection model

(Heckman, 1979) using full maximum likelihood (Greene, 2000; Mad-

dala, 1983). We model auditor choice as a function of firm-specific

variables that have been used in previous studies (Chaney, Jeter, &

Shivakumar, 2004; Choi & Wong, 2007; Fan & Wong, 2005;

Guedhami, Pittman, & Saffar, 2014) and variables that reflect unique

features of the Indonesian setting that are likely to impact auditor

choice. This selection model is estimated simultaneously with the

main regression model.

Our findings are as follows. First, the results of the OLS and the

Heckman treatment effect estimation show a significant negative

effect on the cost of debt from both Big4 and second-tier affiliation.

Second, we find that the effects using the Heckman treatment effect

estimation are larger in absolute value. Third, we find that the effects

on the cost of debt are not only statistically significant but also eco-

nomically relevant. More specifically, the decrease from affiliation

with a Big4 (second-tier) firm in cost of debt is on average 190 (150)

basis points using OLS and 640 (430) basis points using treatment

effect. Fourth, we find no conclusive evidence of a difference in the

magnitude of the effects of Big4 and second-tier affiliation. Although

tests for equality of the effects obtained from the treatment effect

estimation indicate that the effect of Big4 affiliation is larger than the

effect of second-tier affiliation, similar tests for the effects from OLS

estimation show no difference between Big4 and second-tier

affiliation.

Next, we investigate whether the relation between the affiliation

of an audit firm with a Big4 or second-tier audit firm and the effective

interest cost varies with firm characteristics. For lenders, the addi-

tional assurance provided by a high-quality audit firm might be more

pertinent for high risk borrowers. We define subsamples of high‑ver-

sus low-risk borrowers partitioned on the median of the Altman Z-

score (Altman, 1968) and the level of debt, respectively. Our results,

however, show that the negative relation between affiliation with

Big4 and second-tier auditors and the cost of debt cost is independent

from the risk profile of the clients, regardless of how risk is defined.

Next, as an alternative to treatment effect estimation, we control

for endogenous auditor choice by using propensity score matching

(Rosenbaum & Rubin, 1983). We created three pairs of matched sam-

ples: (1) Big4 matched with non-Big4–non-second tier, (2) second tier

matched with non-Big4–non-second tier; and (3) Big4 matched with

second tier. We estimate the propensity scores by including variables

that impact both auditor-selection and the cost of debt. Results of

using multiple regression analysis on the matched samples confirm

that both Big4 affiliation and second-tier affiliation have a significant

negative effect on the cost of debt. These effects are, however,

smaller than those from using treatment effect estimation (i.e., a

reduction in the cost of debt of 190 and 130 basis points for Big4 and

second-tier affiliation, respectively). In line with the OLS estimation

results, there is no evidence that the effects of Big4 affiliation and

second-tier affiliation on the cost of debt differ.

Our work has contributions both from theoretical and practical

points of view. First of all, our results are relevant from the point of

view of managers, who are concerned about their cost of capital in

KURNIAWATI ET AL.

388