Journal of Contemporary Accounting & Economics: Ahsan Habib, Abdul Haris Muhammadi, Haiyan Jiang

Journal of Contemporary Accounting & Economics 13 (2017) 1–19
Contents lists available at ScienceDirect
Journal of Contemporary
Accounting & Economics
journal homepage: www.elsevier.com/locate/jcae
Political connections, related party transactions, and auditor

choice: Evidence from Indonesia q
Ahsan Habib a,⇑, Abdul Haris Muhammadi b, Haiyan Jiang a
a
School of Accountancy, Massey University, Private Bag 102904, Auckland, New Zealand
b
Staff of Directorate of Tax Regulations I, Directorate General of Taxes, Ministry of Finance, Jl. Gatot Subroto Kav., 40-42, Jakarta Selatan, Jakarta 12190, Indonesia
a r t i c l e i n f o a b s t r a c t
Article history: This paper investigates how political connections in concert with related party transactions
Received 5 March 2016 (RPTs) determine auditor choice in Indonesia. Our study is motivated by conflicting find-
Accepted 12 January 2017 ings in the literature on whether politically connected firms appoint reputable auditors
Available online 6 February 2017
(Big 4 auditors). On one hand, politically connected firms are less likely to appoint Big 4
auditors if they wish to cover up RPT-related tunneling activities by providing financial
JEL classification: statements that fail to reflect their true economic performance. On the other hand, politi-
G3
cally connected insiders who refrain from self-dealing would prefer higher-quality finan-
M0
M4
cial reporting and, hence, appoint Big 4 auditors. Using data from Indonesia, we find
support for the former. By documenting the role of RPTs as a motivating factor for politi-
Keywords: cally connected firms to choose non-Big 4 auditors, we enrich the political connection
Political connections and auditor choice literature.
Related party transactions (RPTs) Ó 2017 Elsevier Ltd. All rights reserved.
Indonesia
Auditor choice
1. Introduction
This paper investigates how political connections in concert with related party transactions (RPTs) determine auditor
choice in Indonesia. Competing arguments exist regarding auditor choice by politically connected firms. On one hand, con-
nected firms might prefer to appoint non-Big 4 auditors to mask their tunneling and rent-seeking activities as manifested in
their financial statements. Such tunneling and rent seeking activities are undertaken by politically connected firms in order
to establish and maintain their political connections (Morck et al., 2000; Ma et al., 2013). Given the value of political
connections (e.g. Fisman, 2001), politically connected firms may manipulate accounting numbers to conceal true economic
performance, in order to ensure that their diversionary practices remain a secret (Guedhami et al., 2014). Appointing non-Big
4 auditors is a step towards that direction.
On the other hand, connected insiders who refrain from self-dealing would prefer to choose Big 4 auditors, so that outside
investors will value their reporting transparency positively. It is well accepted that investors value accounting transparency
in order to identify managerial misconduct detrimental to their value maximization interests (Dyck and Zingales, 2004).
Following the arguments in the literature that Big 4 auditors make financial statements more credible (DeFond and
q
We thank an anonymous reviewer for providing numerous constructive comments on the manuscript. We also thank the Editor, Ferdinand Gul for
comments.
⇑ Corresponding author.
E-mail addresses: a.habib@massey.ac.nz (A. Habib), abdulharis.muhammadi@gmail.com (A.H. Muhammadi), hjiang@massey.ac.nz (H. Jiang).
http://dx.doi.org/10.1016/j.jcae.2017.01.004
1815-5669/Ó 2017 Elsevier Ltd. All rights reserved.
2 A. Habib et al. / Journal of Contemporary Accounting & Economics 13 (2017) 1–19
Zhang, 2014), it can be argued that connected firms would prefer appointing Big 4 auditors. In line with this idea, Fan and
Wong (2005) claim that high quality auditors might be hired by the management to increase the credibility of their financial
reports in emerging markets.
Firms with incentives to tunnel resources from minority shareholders require channels through which this can be
achieved. We consider RPTs as one such tunneling channel, and investigate whether connected firms with certain types
of RPTs differ in terms of auditor choice, compared to their non-connected counterparts. A related party transaction is a
transfer of resources, services, or obligations between related parties, regardless of whether or not a price is charged (Inter-
national Accounting Standards 24.9) (IASB, 2009), where a related party is a person or entity related to the entity preparing
its financial statements. These transactions are diverse, and often complex, business transactions between a firm and its own
managers, directors, principal owners or affiliates, and are reasons for concern because they violate arm’s-length market
transaction principles. There is abundant empirical evidence in the extant literature that supports the opportunistic incen-
tive for engaging in RPTs, in particular, RPT loan and loan guarantees (Jian and Wong, 2010; Kohlbeck and Mayhew, 2010;
Chen et al., 2011; Ryngaert and Thomas, 2012; Ying and Wang, 2013).1 Since it is easier for politically connected firms to con-
duct RPTs, because of their large numbers of affiliates with complex interrelationships, we propose that politically connected
firms undertaking opportunistic RPTs would hire non-Big 4 auditors.
However, an alternative argument could be advanced from the supply side, i.e., auditors’ assessment of the audit risk
associated with politically connected firms. Auditors’ evaluation of engagement risk is aimed at minimizing the auditor’s
exposure to litigation loss, loss of reputation, as well as regulation risk (Knechel et al., 2007). Since selection of clients by
auditors is non-random, good quality auditors might avoid auditing risky clients, in our case, politically connected firms,
because of their strong incentives for engaging in tunneling and rent-seeking activities. This perspective suggests that audi-
tors’ choice of clients might drive the results, instead of the discretionary appointment of auditors by politically connected
firms. Since Big 4 auditors are of high quality, they should be more concerned about their clients’ opportunistic business
transactions, including opportunistic RPTs, and may choose to avoid clients with political connections. This argument might
also suggest a negative association between political connections and auditor choice. However, empirical evidence from
Malaysia suggests that auditors charge higher audit fees for politically connected firms instead of resigning from the engage-
ment (Gul, 2006).
Guedhami et al. (2014), an important related paper published recently, find that politically connected firms are more
likely to appoint Big 4 auditors compared to their non-connected peers. We, on the other hand, propose that politically con-
nected firms have incentives to appoint non-Big 4 auditors. This is due to our consideration of firm-level RPTs that encourage
politically connected firms to siphon resources from minority shareholders. Since Guedhami et al. (2014) did not investigate
the possibility that connected firms might engage in rent-seeking using opportunistic RPTs, our prediction of a negative asso-
ciation is justified. That is, Guedhami et al. (2014) considered the incentives for connected firms to be transparent, whereas
we propose that connected firms might extract rents and, hence, choose non-Big 4 auditors.
Indonesia offers an interesting setting to explore this research due to its unique institutional features. First, political con-
nections affect firm value in Indonesia (Fisman, 2001). By using an event study of rumors of the health of former President
Suharto during 1995 to 1997, Fisman (2001) finds that returns of shares having a close relationship with Suharto lost value
compared to those having no affiliation with the Suharto government. In addition, Leuz and Oberholzer-Gee (2006) docu-
ment that the performances of politically connected firms in Indonesia fluctuate following the fortunes of their connections.
Second, Indonesia has a high ownership concentration (Claessens et al., 2000; Brown, 2006), with an average 16.6% of market
capitalization being confined in the hands of a single family: Sudono Salim, a closed ally of Suharto (Brown, 2006). This num-
ber rises to a staggering 57.7% for the top ten families in Indonesia, centered on Suharto political and military connections
(Claessens et al., 2000; Brown, 2006). Such a high ownership concentration gives rise to a type II agency problem, i.e., the risk
of expropriation of minority shareholders’ resources by their controlling owners. Third, RPTs are significant in Indonesia, as
more than 90 per cent of listed firms in Indonesia conduct various forms of RPTs (Utama and Utama, 2013).2 In addition,
Indonesian listed companies are part of complex economic groups with interlocking directorships, reciprocal ownership
arrangement and excessive cross-ownerships, which expedites RPTs. Although the Capital Market and Financial Institutions
Supervisory Agency requires listed firms to disclose RPTs, companies’ complex ownership structures and lack of transparent dis-
closures makes the monitoring of RPTs difficult (Wulandari and Rahman, 2005). Habib et al. (forthcoming) find that politically
connected firms conduct more opportunistic RPTs compared to their non-connected counterparts and further engage in income-
increasing earnings management to mask such opportunistic RPTs. This evidence corroborates politically connected firms’
incentives to appoint non-Big 4 auditors in Indonesia.
For our study, a firm-year observation is categorized as politically connected if at least one of its large shareholders (hav-
ing control of at least 10 per cent of voters directly or indirectly), or its board of directors or board of commissioners, is a
1
For example, Kohlbeck and Mayhew (2010) document a detrimental effect of disclosed RPTs on firm valuations and subsequent returns when compared
with non-RPT firms.
2
The Indonesian Capital Market Supervisory Agency (ICMSA), requires listed firms to announce RPTs to the public as well as to the ICMSA no later than two
working days after the transactions are undertaken if the RPT has a value larger than 0.5% of the firm’s paid capital and greater than IDR 5,000,000,000 (US
$509,840). Hence, most of individual RPTs are arranged below these thresholds in order to avoid public announcement (Utama and Utama, 2014). However,
RPTs having a value less than 0.5% of the firm’s paid capital still need to be reported to the ICMSA and be disclosed in the notes to financial statements (See
Appendix B for further details on RPT regulations in Indonesia).
A. Habib et al. / Journal of Contemporary Accounting & Economics 13 (2017) 1–19 3
current or former (a) member of the parliament, (b) minister or head of a local government, or (c) closely related to a politi-
cian or party. We further decompose political connections into one of three mutually exclusive categories: namely, govern-
ment connections, military connections, and Suharto connections.
Using a panel data of 1428 RPT firm-year observations from 2007 to 2013, we find that politically connected firms with
RPTs in Indonesia tend to choose non-Big 4 auditors. This negative association is more pronounced for firms with govern-
ment connections as opposed to firms with military connections. Firms with Suharto connections, on the other hand, tend
to choose Big 4 auditors. Our results remain robust to controls for a potential self-selection problem.
We extend the auditor choice research by investigating how the choice of auditors in Indonesia is systematically affected
by firms’ political connections (See Appendix A for a discussion on the external auditing environment in Indonesia). We fur-
ther contribute to the auditor choice literature by documenting the important role played by firm-level RPTs. This latter find-
ing is particularly insightful as this provides a contextual explanation for why an opposite result to that of Guedhami et al.
(2014) can be expected. Finally, we enrich the political connection literature, as applied in auditing, by examining the effects
of three mutually exclusive categories of political connections separately. We find that firms having connections with the
government are more inclined to choose non Big 4 auditors.
The remainder of the paper proceeds as follows. Section 2 reviews the relevant literature and develops hypotheses. Sec-
tion 3 describes research design followed by sample selection and descriptive statistics in Section 4. The following section
explains main test results and Section 6 concludes the paper.
2. Literature review and hypotheses development
The classic agency problem between shareholders and managers gives rise to the hiring of auditors, who provide inde-
pendent assurance to corporate stakeholders that financial statements prepared by corporate managers comply with gener-
ally accepted accounting principles (GAAP) (Watts and Zimmerman, 1983). Auditing also plays a significant role in enforcing
and protecting investors’ rights, by detecting expropriation by insiders (Newman et al., 2005), and benefits management by
signalling the reliability of management-provided financial information. A firm’s decision to appoint a certain type of auditor
is, therefore, a crucial element of the auditing landscape.
Previous research on the possible determinants of auditor choice (both brand name and industry specialist auditors)
include culture (Hope et al., 2008), firm-level governance, e.g., ownership structure, country-level investor protection (Fan
and Wong, 2005; Wang et al., 2008; Guedhami et al., 2009; He et al., 2014), managerial incentives (Chen et al., 2015),
and political connections (Guedhami et al., 2014). Our study belongs to the ‘political connection’ domain.
Many benefits accrue to politically connected firms: preferential access to lenders (Johnson and Mitton, 2003; Khwaja and
Mian, 2005; Faccio, 2006; Leuz and Oberholzer-Gee, 2006; Boubakri et al., 2012a); low cost of debt and equity (Houston
et al., 2014); high likelihood of being bailed out (Faccio et al., 2006); profitable government contracts (Goldman et al.,
2009); favorable regulations (Goldman et al., 2009); less monitoring and oversight (Faccio, 2006); lower taxes (Faccio,
2006, 2010); and preferential import licenses and tariffs (Goldman et al., 2009). On the other hand, political connections
are also viewed as harmful to the minority shareholders, as these connections can lead to rent-seeking activities (Frye
and Shleifer, 1997; Faccio, 2006; Boubakri et al., 2012b), tunneling (Qian et al., 2011), and earnings management (Chaney
et al., 2011).
In considering whether to choose a Big 4 auditor, controlling shareholders will assess potential benefits derived, and costs
incurred, as a result of their choice (He et al., 2014). Controlling shareholders might voluntarily adopt bonding mechanisms
in dealing with adverse pricing and a high cost of capital caused by asymmetric information and illiquidity (Jensen and
Meckling, 1976). Fan and Wong (2005) claim that external independent auditors might be hired as monitors, or as one of
the bonding mechanisms, designed to alleviate information and agency problems. Big 4 auditors might offer high quality
audit because they have better monitoring capability (Watts and Zimmerman, 1983), are keen to maintain their reputation,
and are subject to heightened litigation (Hope et al., 2008; Guedhami et al., 2014). In addition, Big 4 auditors might deliver a
high quality of assurance consistently, owing to their global operation (Humphrey et al., 2009; Guedhami et al., 2014). For
politically connected firms, being audited by Big 4 auditors can, therefore, reap benefits, such as better transparency, higher
valuation, lower earnings management, and cheaper cost of capital (Guedhami et al., 2014; He et al., 2014).
On the other hand, the appointment of a Big 4 auditor may be costly, since controlling shareholders have fewer oppor-
tunities for expropriation owing to the significant role of Big 4 auditors in monitoring financial reporting discretion of con-
trolling shareholders (Guedhami et al., 2009). Politically connected firms expecting to obtain benefits from their allies, tend
to appoint non-Big 4 auditors for the following reasons. First, connected firms are inclined to render more opaque financial
statements in order to conceal their tunneling and rent seeking activities (He et al., 2014; Chen et al., 2015). Second, any
increase in transparency will decrease the ability of controlling shareholders and their political allies to enjoy the private
benefits of control (Leuz and Oberholzer-Gee, 2006; Piotroski et al., 2015). For politically connected firms, transparency is
costly, since it might result in unwanted scrutiny, restricting the possibility of exploiting weak corporate governance
(Bona-Sánchez et al., 2014; Piotroski et al., 2015). Further, He et al. (2014) claim that when controlling shareholders develop
political connections, their political partners need more secrecy, so that their reputations are maintained at the expense of
financial reporting transparency. Third, the lack of transparency of non-Big 4 auditing does not reduce their chance of getting
easy credit from state-owned banks anyway (Dinç, 2005; Guedhami et al., 2014). Supporting this idea, Bushman et al. (2004)
claim that in return for bribes, nepotism, and other political supports, the politicians might exercise their control over state
owned banks to provide preferential financing to their connected firms. Financial reporting transparency may have a min-
imal role in this setting, diminishing the incentive for appointing a more expensive Big 4 auditor. Moreover, Leuz and
Oberholzer-Gee (2006) find that Indonesian firms with close connections to the state avoid raising capital from abroad with
more onerous disclosure requirements that include, but are not limited to, Big 4 audits of the financial statements.3
We explore this underlying tension, and investigate whether connected firms appoint high or low quality auditors. The
preceding discussion suggests that politically connected firms weigh the marginal benefits of appointing a Big 4 auditor
against its marginal costs. The importance of political connections to the demand for Big 4 audits is an open empirical ques-
tion that depends on which financial reporting incentive (transparency versus extraction of private benefits) dominates. We,
predict that firms with political ties tend to hire low quality auditors to tunnel resources and engage in rent-seeking activ-
ities (as developed in H2 below).
H1. Politically connected firms are less likely to appoint Big 4 auditors.
However, a supply side argument concerning auditor choice of clients may also be relevant in understanding the associ-
ation between political connection and auditing in Indonesia. Because of the non-random selection of clients by auditors,
good quality auditors might avoid auditing risky clients, in our case, politically connected firms, because of their tunneling
and rent-seeking activities. An extensive body of literature supports the view that Big 4 auditors provide higher audit quality
because of their exposure to greater litigation risk (‘‘deep pocket” theory) and greater reputation risk (independence con-
cern) (DeFond and Zhang, 2014). This argument suggests that Big 4 auditors should be more concerned about their clients’
opportunistic business transactions, and may choose to avoid risky clients to minimize their exposure to litigation and rep-
utational losses. However, high quality auditors also tend to be larger, with the capacity to diversify client risk and, thus, are
better able to bear the risk of such clients. Hence it is not obvious whether high quality auditors avoid risky clients (DeFond
and Zhang, 2014). Empirical evidence, too, finds support in favor of retaining politically connected firms, but charging a fee
premium (Gul, 2006).
In Indonesia, connected firms can be classified further into government, military, and Suharto connections. Our sample
covers two consecutive periods of Susilo Bambang Yudhoyono (SBY)’s presidency, from 2004 to 2014. With respect to gov-
ernment connection, extant literature, argues that government plays a key role in controlling and allocating key resources
(Child, 1994; Li et al., 2008). Firms willing to maintain an ongoing relationship with government need to share the rents
extracted through expropriation of minority resources and, as well, obfuscate their financial reports to mask tunneling activ-
ities. Appointing a non-Big 4 auditor is a proactive decision to accomplish this.
H1A. Firms having political connections to government are more likely to hire non-Big 4 auditors.
On the other hand, Suharto-connected firms are more likely to appoint Big 4 auditors. When Suharto was in power, firms
having an affiliation with his regime through his families, friends, and military connections enjoyed ample privileges (Brown,
2006), e.g., preferential loans from state owned banks through memo-lending and exclusive import licenses (Leuz and
Oberholzer-Gee, 2006). However, after Suharto’s resignation, firms having connections with Suharto had reduced access
to government officials. They had difficulties in establishing a connection with the new government, and experienced loss
of government contracts, distributorships, and brokerage monopolies (Fukuoka, 2013).
With respect to military influence in Indonesia, it has been observed that, during the Suharto regime both active and for-
mer military personnel held strategic posts at the national and regional level, including managerial positions in state owned
enterprises (Brown, 2006; Bhakti et al., 2009; Sebastian and Iisgindarsah, 2013). Suharto handed over state owned enter-
prises, previously seized from Dutch companies, to be managed by military personnel. However, with the end of the Suharto
era, foundations belonging to the military, Suharto’s family and Golkar were under investigation (Brown, 2006). Therefore,
Mietzner (2006) concludes that the army have lost formal political influence considerably, and they do not serve as a back-
bone for the incumbent regime anymore. With these benefits gone, firms having Suharto as well as the military connections
had less incentive to engage in tunneling and financial report manipulation in order to obfuscate such tunneling. Based on
these arguments we hypothesize the following:
H1B. Firms having Suharto as well as military connections are more likely to hire Big 4 auditors compared to firms with
government connections.
Note that H1 is opposite to what Guedhami et al. (2014) established. This is due to our consideration of firm-level RPTs
that encourage politically connected firms to siphon resources from minority shareholders. Since Guedhami et al. (2014) did
not investigate the possibility that connected firms might engage in rent-seeking using opportunistic RPTs, our prediction of
a negative association is justified. That is, Guedhami et al. (2014) considered the incentives for connected firms to be trans-
parent, whereas we propose that connected firms might extract rents and, hence, choose non-Big 4 auditors.
3
Another perspective holds that politically connected top managers could give outside investors an impression that the government favors the connected
firms, which could conceivably be more effective than hiring high-quality auditors. In such a case, outside investors would pay less attention to audit quality, so
that the firms would be less motivated to hire high-quality auditors. Investors might also agree that building political connections is more cost-effective than
hiring reputable auditors.
Our next hypothesis explicitly considers the mechanisms through which connected firms might conduct tunneling activ-
ities, and whether that had a bearing on auditor choice by connected firms. We propose that RPTs is one such channel
through which politically connected firms might conduct resource diversion. RPTs are used to structure transactions, e.g.,
tunneling, propping or earnings management, among their affiliates in order for insiders to expropriate minority resources
(Thomas et al., 2004; Cheung et al., 2009). In the context of Indonesia, Habib et al. (forthcoming) document that politically
connected firms conduct more RPTs compared to their non-connected counterparts, and that this effect is more pronounced
for firms with government connections. Further analysis reveals that the connected firms use RPTs to tunnel resources, and
engage in income-increasing earnings management to mask such tunneling activities. Since it is easier for politically con-
nected firms to conduct RPTs owing to the presence of a large number of affiliates with complex inter-relationships among
them, we propose that politically connected firms with opportunistic RPTs would hire non-Big 4 auditors.
Extant literature has documented that opportunistic RPTs are primarily conducted through RPTs involving loan guaran-
tees and capital transfers (RPLOAN). Prior literature reveals that RP loans and guarantees have been used by parent compa-
nies for tunneling or siphoning resources out of their listed subsidiaries (Berkman et al., 2009; Jiang et al., 2010). Empirical
evidence shows that, compared to those with low levels of RP loans and guarantees, Chinese firms with high levels of RP
loans and guarantees demonstrate significantly worse future performances including sharp declines in profitability, and a
higher likelihood of entering financial distress in the future (Jiang et al., 2010). Habib et al. (forthcoming) find that connected
firms use primarily RP loans and guarantees to tunnel resources in Indonesia. RP loans are generally not made as part of the
normal course of business, and most loans do not accrue interest. If RPTs involving loans allow politically connected firms to
siphon resources, then there is an incentive for those firms to manipulate financial reports in order to obfuscate true eco-
nomic performance. This argument suggests that politically connected firms with RPT loans are more likely to choose
non-Big 4 auditors. It is important to note that this argument does not imply that non-connected firms don’t use RPT loans
for opportunistic reasons. However, we do expect the effect to be more pronounced for connected firms than for their non-
connected counterparts. The following hypothesis is developed:
H2. Firms with RP loan and guarantees are less likely to appoint Big 4 auditors and this is more pronounced for politically
connected firms.
Because we hypothesize that politically connected firms conduct opportunistic RPTs for rent-seeking activities, the pre-
ceding hypotheses are valid in the context of such RPTs.
3. Research design – empirical model
To test H1, we develop the following regression model:

AUDITOR ¼ c0 þ c1 PCON þ c2 OWNCON þ c3 FOWN þ c4 SIZE þ c5 LEV þ c6 GROWTH þ c7 INV þ c8 SEGMENT
þ c9 FINANCE þ c10 ROA þ c11 AC þ c12 IC þ c13 jDACj þ IndustryFE þ YearFE þ e ð1Þ
where AUDITOR is dummy variable, coded 1 for Big 4 auditors and zero otherwise. The local Indonesian audit firms that are
affiliated with the Big 4 audit firms are: Tanudiredja, Wibisana and Rekan (PWC); Purwantono, Suherman and Surja (EY);
Osman Bing Satrio and Rekan (Deloitte); and Siddharta Siddharta and Widjaja (KPMG). PCON is an indicator variable coded
1 if the sample observations have political connections, 0 otherwise. We expect a negative and significant coefficient on
PCON to suggest that politically connected firms will choose non-Big 4 auditors.
We include a set of control variables based on prior literature on the determinants of auditor choice (DeFond et al., 2000;
Wang et al., 2008; Guedhami et al., 2009, 2014). Larger firms (SIZE), firms with growth opportunities (GROWTH), firms with
more inventories in their balance sheet (INV), multi-segment firms (SEGMENT), new insurance of finance (FINANCE), prof-
itable firms (ROA), firms with larger foreign ownership (FOWN), and less-leveraged (LEV) firms are more likely to appoint
a Big 4 auditor. FOWN is the total percentage of shares owned by foreign institutional investors. Foreign institutional inves-
tors prefer Big 4 auditors to ensure the quality and the credibility of financial statements (Guedhami et al., 2009; He et al.,
2014). The association between ownership concentration (OWNCON) and auditor choice is also expected to be positive (Fan
and Wong, 2005).
SIZE is measured as the natural log of total assets. GROWTH is the market value of equity divided by the book value of
equity. ROA is net income divided by total assets. LEV is the total long-term debt divided by total assets. AC is dummy vari-
able coded 1 if the firm has established an audit committee, and zero otherwise. We expect a positive coefficient on AC, since
independent audit committees demand higher quality audits (Carcello et al., 2002; Abbott et al., 2003). IC is dummy variable
coded 1 for a firm having independent commissioners, and zero otherwise.4 We expect a negative and significant coefficient
4
Indonesia adopts a two-tier system for board structures: the board of directors (BODs) and the board of commissioners (BOCs). The BODs serve as firm’
executives whereas the BOCs have responsibility to supervise management policies and to advice the BODs. The role of the BOCs in the two-tier system is
comparable to that of the BODs in a one tier system. In order to improve corporate governance in Indonesia, the Capital Market and Financial Institutions
Supervisory agency requires listed firms to have independent commissioners and audit committees. However, those boards did not carry out their legal duties
properly because both BOCs and BODs represent the interests of, and are often selected by the15 families who control the majority of the listed companies
(Wulandari and Rahman, 2005).
on discretionary accruals (DAC) to suggest that the firms having higher DAC will prefer to appoint non-Big 4 auditors in order to
convey less information on their DAC. Appendix C provides the variable definitions.5
To estimate DAC we use the cross-sectional modified Jones model, controlling for firm performance (Kothari et al., 2005).
We estimate the following model for all firms in the same industry (using the SIC two-digit industry code) with at least eight
observations in an industry in a particular year:
ACC t =TAt1 ¼ c0 ð1=TAt1 Þ þ c1 ½ðDSALESt DRECEIVABLEt Þ=TAt1 þ c2 ðPPEt =TAt1 Þ þ c3 ðROAt1 Þ þ et ð2Þ
where ACC is total accruals, calculated as earnings before extraordinary items and discontinued operations minus operating
cash flows; TA is total assets in year t-1; DSALES is change in sales from year t-1 to year t; DRECEIVABLE is change in accounts
receivable from year t-1 to year t; PPE is gross property plant and equipment; ROA is the prior year’s return-on-assets mea-
sured as earnings before extraordinary items and discontinued operations, divided by total assets for the previous year. The
coefficient estimates from Eq. (2) are used to estimate the non-discretionary component of total accruals (NDAC) for our sam-
ple firms. The discretionary accruals are then the residual from Eq. (2), i.e. DAC = ACC-NDAC.
An interesting aspect of the political connection landscape in Indonesia relates to the different types of connection, which
are not captured in a single PCON variable. We categorize three mutually exclusive types of political connection: GCON
(dummy variable, 1 for government connected firms, 0 otherwise); MCON (dummy variable, 1 for military connected firms,
0 otherwise); and SCON (dummy variable, 1 for Suharto connected firms, 0 otherwise). The regression equation below is,
therefore, estimated:
AUDITOR ¼ c0 þ c1 GCON þ c2 MCON þ c3 SCON þ c4 OWNCON þ c5 FOWN þ c6 SIZE þ c7 LEV þ c8 GROWTH þ c9 INV
þ c10 SEGMENT þ c11 FINANCE þ c12 ROA þ c13 AC þ c14 IC þ c15 jDACj þ YearFE þ IndustryFE þ e ð3Þ
We expect the coefficient on GCON to be negative, and that on MCON and SCON to be positive.
Finally, we develop the following regression model to test for the effects of RPTs on auditor choice, conditional on political
connection:
AUDITOR ¼ c0 þ c1 PCON þ c2 RPT þ c3 PCON RPT þ c4 OWNCON þ c5 FOWN þ c6 SIZE þ c7 LEV þ c8 GROWTH
þ c9 INV þ c10 SEGMENT þ c11 FINANCE þ c12 ROA þ c13 AC þ c14 IC þ c15 jDACj þ YearFE þ IndustryFE þ e ð4Þ
We use three categories of RPTs. RPT_RATIO is ratio of gross RPT over total assets. Deflation by assets is justified since RPTs
vary with firm size. OPRPT_RATIO is operating RPTs deflated by total assets. Finally, LOAN_RATIO is RPT loan deflated by total
assets. Operating RPTs, dealing with sales and purchases of goods and services, consist mainly of trade relationships. From an
efficiency-enhancing perspective, operating RPTs are argued to be used within corporate groups as a way of optimizing inter-
nal resource allocation, reducing transaction costs (Jian and Wong, 2010; Chen et al., 2012) and improving return-on-assets
(Khanna and Palepu, 2000). This is in contrast to RPT loan which is primarily a mechanism for siphoning resources. Our vari-
able of primary interest is the interactive variable PCON⁄RPT. If politically connected firms make use of RPTs to siphon
resources (LOAN_RATIO), then we would expect a negative and significant coefficient on PCON⁄LOAN_RATIO. The coefficient
on PCON⁄OPRPT_RATIO is expected to be positive. The coefficient on c2 captures auditor choice by non-connected firms in the
presence of RPTs.
4. Sample selection and descriptive statistics
Data on the number and amount of RPT is hand-collected from audited financial reports downloaded mainly from the
website of the Indonesia Stock Exchange (http://www.idx.co.id/index-En.html). If not available, the data are derived from
the websites of Indonesian listed firms. In addition, the following corporate governance data are also manually collected
from audited financial statements or annual reports: board of directors, board of commissioners, independent commission-
ers, audit committee, the names of audit firms, the names and percentage of share ownerships, and information on repor-
table segments. Financial statement data are collected from the Research Insight-Global Vantage database. Since most of the
data for RPT are in Indonesian Rupiah, the data for RPT are translated into US$ by using the exchange rate available from the
DataStream. Finally, market data are retrieved from DataStream.
The criteria for defining politically connected firms follows Faccio (2006), Chaney et al. (2011), and Guedhami et al. (2014)
with necessary modification to the Indonesian context. A firm-year observation will be categorized as politically connected if
at least one of its large shareholders (having at least 10 per cent direct or indirect voting rights) or its board of directors and
board of commissioners is a current or former (a) member of parliament, (b) minister or head of a local government, or (c) is
closely related to a politician or party. Connection with government ministries is extended to close relatives (spouse, sons or
daughters, and other immediate family relationship). Close relationships with politicians or parties encompass well-known
5
We did not include some other control variables as in Guedhami et al. (2014) for reasons outlined below. Data on FOREIGNSALE is not reported by
companies in their published annual report. We did not include CROSS-LIST because there are only two companies that are cross-listed [Aneka Tambang
(Persero) Tbk and Telekomunikasi Indonesia (Persero) Tbk]. CONTROLRIGHT (the percentage of voting rights belonging to the ultimate owner) and
CASHFLOWRIGHT (the percentage of voting rights belonging to the ultimate owner) are unavailable because listed firms in Indonesia are only obliged to disclose
their legal owners instead of their beneficial or ultimate owners (Utama and Utama, 2013).
friendships as identified by The Economist, Forbes or Fortune; share ownership or a directorship held by former ministers, for-
mer heads of government, former member of parliament and current politicians (Faccio, 2006; Chaney et al., 2011); well doc-
umented relationships with political parties as utilized by Johnson and Mitton (2003); and famous connections adopted by
Fisman (2001) and Leuz and Oberholzer-Gee (2006).
To establish those political connections, the names of boards of directors, boards of commissioners and data of ownerships,
including the name and the percentage of ownership, were gathered from the Indonesia Stock Exchange (http://www.idx.co.
id/index-En.html), company websites, audited financial reports and annual reports. The names of members of parliament
were collected from the website of the Indonesia House of Representatives (http://www.dpr.go.id/id/anggota/), the names
of members of cabinet were gathered from the website of the cabinet secretariat of the Republic of Indonesia (http://set-
kab.go.id/en/profil-kabinet.html). The names of heads of local government (governors) were collected from (http://www.ke-
mendagri.go.id/staff-directory/gubernur-dan-wakil-gubernur). The names of members of parliament, members of cabinet,
and heads of local government were matched with the names of boards of directors, boards of commissioners and sharehold-
ers. In addition, political connections could also be identified from the profiles of members of boards of commissioners and
directors described in the annual reports. Our primary variable of interest, PCON, is an indicator variable coded 1 for the
firm-year observation fulfilling at least one of the abovementioned criteria, and 0 otherwise. We further categorize politically
connected firms into three mutually exclusive categories: namely, government connections, military connections and Suharto
connections.
The sample period is from 2007 until 2013, which covers two periods of the first directly elected president, Susilo Bam-
bang Yudhoyono (SBY)’s presidency, i.e., 2004–2014. The first period of President SBY’s administration was from 20 October
2004 until 20 October 2009 and the second period was from 20 October 2009 until 20 0ctober 2014. The financial year: 2014,
is excluded from the observation owing to a political regime change whereby a newly elected government took power.
Table 1, Panel A illustrates our sample selection process. The number of listed non-financial firms on the Indonesian Stock
Exchange in 2004 was 244 firms. This number increased to 405 firms in 2013 giving us a total initial sample of 3149 firm-
year observations for the period 2004 to 2013 inclusive. We deleted 772 firm-year observations pertaining to the 2004 to
2006 sample period because we could retrieve valid data for only 25 firm-year observations.6 From the remaining sample
of 2377 firm-year observations we deleted 481 firm-year observations with unavailable audit reports. This result in a total
1896 firm-year observations for matching with other variables required to run a regression on RPTs. We then deleted 113
firm-year observations with zero RPT values and 85 firm-year observations with negative equity values. Finally, we deleted
a further 269 firm-year observations with missing data on the relevant control variables, resulting in a final usable sample
of 1429 firm-year observations for the period 2007 to 2013.
Panel B, Table 1 presents descriptive statistics for the variables used in the regressions. We winsorize the continuous vari-
ables at 1% and 99% of their respective distributions to control for the effects of outliers. About 43% of the firm-year obser-
vations are audited by a Big 4 audit firm. Thirty nine percent of the observations have political connections, split among
GCON (24%), MCON (12%) and SCON (3%). The average RPT_RATIO (Gross RPT/Total assets) is 0.50 of total assets. The corre-
sponding values for OPRPT_RATIO (Gross Operating RPTs/Total assets) and LOAN_RATIO (Gross RPT_LOAN/Total assets) are
0.28 and 0.11 respectively.
These two categories do not add up to 0.50 because there are RPTs that do not belong to operating and loan RPTs. How-
ever, we do not include other RPTs in our regression specification because of a lack of theoretical prediction regarding its
impact on auditor choice by politically connected firms. Sample firms have growth opportunities (mean GROWTH is 3.17),
are low-levered (an average of 0.13), and profitable (average ROA of 7%). Eighty one percent of the sample observations have
established an audit committee.
The industry distribution of sample companies is presented in Panel C, Table 1, revealing that materials account for
23.79% of the total sample observations, followed by consumer discretionary and industrials with 20.08% and 17.35% of sam-
ple observations respectively. Finally, Panel D presents a univariate test of the difference in means for the variables between
connected and non-connected firms. The proportion of firm years being audited by Big 4 firms is much higher for connected
as opposed to non-connected firms (an average of 0.53 versus 0.37, t-stat of difference in means is 5.53). This is contrary to
our hypothesized negative association between PCON and AUDITOR. Politically connected firms are larger, more levered,
higher growth firms. There is no significant difference in gross RPT ratios between connected and non-connected firms
although OPRPT_RATIO is significantly larger for the non-connected compared to their connected firm counterparts (t-stat
of difference in means is 1.72, p < 0.10).
Correlations among the variables for auditor choice are presented in Table 2. The correlation between PCON and AUDITOR
is positive and significant at better than the 1% level. This is contrary to our hypothesized negative association between PCON
and AUDITOR. Most of the independent variables are correlated positively with the dependent variable, AUDITOR, at better
than the 1% level. Untabulated correlation reveals that MCON and SCON are positively correlated with AUDITOR, while GCON
is negatively correlated.
6
We contacted the Financial Services Authority (FSA), formerly known as the Indonesian Capital Market and Financial Institution Supervisory Agency,
regarding the availability of hard copy annual reports. Since 05 July 2011, the FSA requires listed firms to submit both hard copy and soft copy audited financial
reports. However, the FSA does not allow public access to those audited financial reports. The Indonesia Stock Exchange also used to receive hard-copies of
audited financial reports, but those are kept in storage outside of Jakarta. Currently, it maintains only soft-copies that can be downloaded from its website: a
procedure that we followed in collecting more recent annual reports.
Table 1
Sample selection procedure and descriptive statistics.
Panel A: Sample selection procedure

Sample selection process Observations
Number of non-financial firm-year observations from 2004 to 2013 3149
Less: Firm year observations pertaining to 2004–2006 sample period dropped because of too few available annual reports (772)
Less: Firm year observations with unavailable audit reports during 2007–2013 period (481)
Less: Firm year observations with zero RPT values (113)
Less: Number of firm-year observations with negative book value (distress firms) (85)
Number of firm-year observations with complete non zero RPT data 1698
Less: missing data of other control variables (269)
Number of firm-year observations for the baseline regression 1429
Panel B: Descriptive statistics

Variables Mean SD 25% Median 75%
AUDITOR 0.43 0.50 0.00 0.00 1.00
PCON 0.39 0.49 0.00 0.00 1.00
GCON 0.24 0.43 0.00 0.00 0.00
MCON 0.12 0.32 0.00 0.00 0.00
SCON 0.03 0.19 0.00 0.00 0.00
RPT_RATIO 0.50 0.88 0.04 0.21 0.66
OPRPT_RATIO 0.28 0.57 0.00 0.03 0.27
LOAN_RATIO 0.11 0.29 0.00 0.01 0.08
OWNCON 0.71 0.19 0.59 0.74 0.85
FOWN 0.28 0.30 0.00 0.15 0.52
SIZE 19.00 1.67 17.89 18.99 20.15
LEV 0.13 0.15 0.00 0.08 0.21
GROWTH 3.17 5.52 0.94 1.66 3.28
INV 0.16 0.16 0.02 0.12 0.24
SEGMENT 0.90 0.53 0.69 1.1 1.39
FINANCE 0.14 0.34 0.00 0.00 0.00
ROA 0.07 0.12 0.01 0.05 0.11
AC 0.81 0.39 1.00 1.00 1.00
IC 0.92 0.27 1.00 1.00 1.00
|DAC| 0.10 0.10 0.03 0.07 0.13
Panel C: Industry distributions

Sector code Economic sector description Observations % of observations
1000 Materials 340 23.79
2000 Consumer discretionary 287 20.08
3000 Consumer staples 222 15.54
3500 Health care 42 2.94
4000 Energy 41 2.87
5000 Real estate management & development 168 11.76
6000 Industrials 248 17.35
8000 Information technology 40 2.80
8600 Telecommunication service 41 2.87
1429 100.00
Panel D: Univariate test
Variables PCON = 1 [n = 562] PCON = 0 [n = 867] t-test of difference in mean
AUDITOR 0.53 0.37 5.53***
RPT_RATIO 0.47 0.51 1.02
OPRPT_RATIO 0.24 0.30 1.72⁄
LOAN_RATIO 0.11 0.12 0.42
OWNCON 0.51 0.50 0.68
FOWN 0.23 0.31 4.55***
SIZE 20.03 18.32 21.95***
LEV 0.16 0.11 5.23***
GROWTH 3.73 2.81 3.08***
INV 0.14 0.18 4.32***
SEGMENT 3.06 2.55 7.29***
FINANCE 0.14 0.13 0.75
ROA 0.08 0.06 1.88⁄
AC 0.89 0.76 6.00***
IC 0.91 0.92 0.57
|DAC| 0.09 0.10 0.99
**
Note: Variable definitions are in Appendix C. is significant at p < 0.05 respectively (two-tailed test).
***
Significant at p < 0.01.
A. Habib et al. / Journal of Contemporary Accounting & Economics 13 (2017) 1–19
Table 2
Correlation analysis.
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17)
AUDITOR (1) 1.00
PCON (2) 0.15 1.00
RPT_RATIO (3) 0.11 0.03 1.00
LOAN_RATIO (4) 0.14 0.05 0.33 1.00
OPRPT_RATIO (5) 0.17 0.01 0.59 0.11 1.00
OWNCON(6) 0.04 0.11 0.49 0.08 0.12 1.00
FOWN (7) 0.18 0.15 0.04 0.17 0.02 0.06 1.00
SIZE (8) 0.16 0.12 0.04 0.14 0.00 0.03 0.30 1.00
LEV (9) 0.40 0.50 0.02 0.01 0.07 0.11 0.21 0.09 1.00
GROWTH (10) 0.00 0.13 0.05 0.16 0.15 0.08 0.14 0.01 0.27 1.00
INV(11) 0.10 0.08 0.02 0.01 0.05 0.07 0.02 0.02 0.01 0.00 1.00
SEGMENT(12) 0.00 0.12 0.09 0.11 0.05 0.10 0.06 0.02 0.09 0.30 0.06 1.00
FINANCE(13) 0.05 0.14 0.04 0.00 0.05 0.02 0.17 0.12 0.20 0.02 0.06 0.03 1.00
ROA (14) 0.02 0.02 0.06 0.02 0.08 0.04 0.00 0.01 0.04 0.17 0.08 0.13 0.01 1.00
AC (15) 0.22 0.04 0.05 0.08 0.16 0.13 0.09 0.01 0.17 0.28 0.20 0.05 0.02 0.24 1.00
IC (16) 0.13 0.16 0.06 0.06 0.12 0.08 0.09 0.06 0.23 0.06 0.01 0.07 0.03 0.04 0.08 1.00
|DAC| (17) 0.05 0.01 0.06 0.03 0.01 0.06 0.05 0.05 0.11 0.07 0.05 0.04 0.02 0.05 0.01 0.20 1.00
Note: Italicized and bold-faced correlations are significant at p < 0.01. Variable definitions are in Appendix C. The correlation is based on a full sample of 1429 firm-year observations.
9
5. Main test results
5.1. Political connections and auditor choice
In a multivariate regression framework, we estimate the impact of political connections on the likelihood that firms
will hire a Big 4 auditor to examine the prediction in H1. Next, we analyze the prediction in H2 that RPTs will moderate
the association between political connections and auditor choice. Table 3, Panel A presents the regression results for the
association between political connections and auditor choice (Big 4 versus non-Big 4) (Columns 1 and 2) and the effect
on that association of gross RPTs (Columns 3 to 5). We control for industry and year fixed effects in all the regression
models. We also report all regression results with clustered standard errors at the firm level (Petersen, 2009; Gow et al.,
2010).
Our baseline regression in Column (1) reveals that the coefficient on PCON is negative and significant (coefficient 0.363,
z-statistic 2.22, p<0.05), suggesting that politically connected firms are less likely to hire Big 4 auditors compared to their
non-connected counterparts. In terms of economic significance, the coefficient estimate of 0.363 means that political affil-
iations decrease the likelihood of appointing a Big 4 auditor by 5.8% with all other variables are assigned their mean values.7
The negative coefficient could be consistent with the argument that politically connected firms avoid Big 4 auditors in order to
conceal tunneling activities knowing that high quality auditors are competent to detect. This supports H1. However, this neg-
ative coefficient could also be consistent with auditor’s risk aversion argument, i.e., Big 4 audit firms might avoid auditing risky
clients to mitigate their litigation and reputation losses. We perform additional tests in the sensitivity test section to rule out
this possibility.8
Column (2) reports this result with GCON showing a negative and significant coefficient (coefficient 0.877, z-statistic
4.36, p < 0.01) while that on SCON loads positively (coefficient 2.358, z-statistic 4.05, p < 0.01). In terms of economic sig-
nificance, the coefficient estimate of 0.877 on GCON means that government political affiliations decrease the likelihood
of appointing a Big 4 auditor by about 13% when all other variables are assigned their mean values. These findings are con-
sistent with H1A. However, we interpret the coefficient on SCON cautiously, given the small number of observations. Among
the other firm-level determinants, we find that firm size, ownership concentration, foreign ownership, growth, and the exis-
tence of an audit committee are related to auditor choice positively; while leverage and profitability are related to auditor
choice negatively.
5.2. Political connections, RPTs, and auditor choice
Next, we test our prediction that RPTs play a significant moderating role in the association between political connections
and auditor choice. We test Eq. (4) using RPT_RATIO (Gross RPT/Total assets) and report the result in Column (3). Our variable
of primary interest is the sign and significance of the coefficient on the interactive variable PCON⁄RPT_RATIO (c3). We expect
the coefficient to be negative to support our hypothesis that politically connected firms use RPTs opportunistically to siphon
off resources for their private benefits and, hence, are less likely to choose Big 4 auditors. Results in Column (3) are consistent
with our predictions. We find the coefficient on the standalone variable RPT_RATIO (c3) is negative (coefficient 0.271, z-
statistic 2.21, p < 0.05) as well as that on the interactive variable (1.163, z-statistic 4.12, p < 0.01). It is interesting to
note that the negative and significant coefficient on RPT_RATIO suggests that non-connected firms with RPTs also tend to
appoint non-Big 4 auditors, although the coefficient is much smaller than for their connected counterparts. Columns (4)
and (5) consider the PCON categories, i.e., GCON and MCON, and run Eq. (4) without the interaction variable, while limiting
the sample to PCON = 1 only. The coefficient on RPT_RATIO is negative and significant for GCON only (coefficient 2.417, z-
statistic 4.96, p < 0.01). The coefficient on MCON is insignificant and that on SCON is undetermined because of the small
sample size.
Panel B, Table 3 decomposes total RPTs into operating RPTs and RPT loans and reruns Eq. (4). Columns (1) to (3) report
results for operating RPTs, Columns (4) to (6) for RPT loans, and finally Columns (7) to (9) for both operating RPTs and RPT
loans in the same regression model. The coefficient on OPRPT_RATIO is positive and significant in Column (1) (coefficient
0.47, z-statistic 3.40, p < 0.01) suggesting that firms conducting operating RPTs tend to appoint Big 4 audit firms. The coef-
ficient on the interactive variable PCON⁄OPRPT_RATIO is negative but insignificant (coefficient 0.517, z-statistic 1.16). The
coefficient on OPRPT_RATIO for the GCON and MCON sub-samples is insignificant. Column (4) reveals a marginally significant
negative coefficient on the interactive variable PCON⁄LOAN_RATIO (coefficient 1.373, z-statistic 1.73, p < 0.10). The
negative coefficient implies that politically connected firms conducting RPTs involving intercorporate loans tend to choose
non-Big 4 auditors. This is consistent with H2, that intercorporate loans enable connected firms to transfer resources from
minority shareholders, and in order to conceal such rent-seeking activities, connected firms choose non Big 4 audit firms. The
7
Although we follow an extensive body of prior research that operationalizes audit quality with the presence or absence of a Big 4 auditor, we also wanted to
examine whether our results are sensitive to an alternative definition: firm-level audit fees, following extant research that audit fees are higher when auditors
expend more effort on an engagement, translating into better audits (e.g. Davis et al., 1993; Whisenant et al., 2003). However, we found that Indonesian firms
do not disclose audit fees as a separate line item, and neither do they provide any note disclosing the amount of audit fees paid to the auditor.
8
We also performed a reverse regression technique whereby we regressed PCON on AUDITOR and the remaining control variables. The untabulated result
shows the coefficient on AUDITOR to be negative but insignificant (coefficient -0.10, z-statistic -0.64).
Table 3
Political connections, RPTs, and auditor choice.
AUDITOR ¼ c0 þ c1 PCON þ c2 OWNCON þ c3 FOWN þ c4 SIZE þ c5 LEV þ c6 GROWTHc7 INV þ c8 SEGMENT þ c9 FINANCE þ c10 ROA þ c11 AC þ c12 IC þ c13 jDACj þ Industry þ Year þ e ð1Þ
Panel A: Political connections, RPTs, and auditor choice
Only PCON = 1 Only PCON = 1
RPT_RATIO RPT_RATIO RPT_RATIO
Expected Sign (1) (2) (3) (4) (5)
Baseline Baseline PCON GCON = 1 MCON = 1
Variables Coefficient [z-stat] Coefficient [z-stat] Coefficient [z-stat] Coefficient [z-stat] Coefficient [z-stat]
PCON 0.363** – 0.007 – –
[2.22] [0.04]
GCON – 0.877*** – – –

[4.36]
MCON – 0.193 – – –
[0.88]
SCON + – 2.358*** – – –
[4.05]
RPT_RATIO – – 0.271** 2.417*** 1.16
[2.21] [4.96] [0.78]
PCON⁄RPT_RATIO – – 1.163*** – –
[4.12]
OWNCON + 2.150*** 2.229*** 2.620*** 4.340*** 5.609***
[6.85] [7.03] [8.12] [5.30] [3.63]
FOWN + 1.866*** 1.857*** 1.936*** 4.229*** 5.163***
[7.59] [7.36] [7.32] [6.01] [5.11]
SIZE + 0.805*** 0.840*** 0.906*** 0.731*** 1.427***
[13.02] [12.81] [12.94] [3.37] [3.76]
LEV 1.283** 1.429** 1.606** 3.125 6.753***
[2.03] [2.22] [2.36] [1.63] [2.63]
GROWTH + 0.047** 0.052*** 0.047** 0.049* 0.116
[2.47] [2.74] [2.46] [1.81] [0.93]
INV + 1.225*** 1.441*** 1.062** 3.881** 0.632
[2.60] [3.01] [2.09] [2.24] [0.33]
SEGMENT + 0.049 0.072 0.015 0.022 0.005
[0.36] [0.51] [0.11] [0.05] [0.03]
FINANCE ? 0.341 0.312 0.387 0.066 0.289
[1.49] [1.33] [1.63] [0.13] [0.37]
ROA + 1.984** 1.803** 1.512* 11.362*** 1.338
[2.44] [2.15] [1.76] [3.36] [0.40]
AC + 0.726*** 0.847*** 0.633*** 0.577 4.304***
[3.69] [4.18] [3.28] [0.78] [4.38]
IC + 0.066 0.262 0.184 0.101 0.524
[0.26] [1.11] [0.72] [0.18] [0.45]
|DAC| ? 0.123 0.090 0.035 1.352 7.542*
[0.15] [0.11] [0.04] [0.48] [1.79]
Constant 16.710*** 17.614*** 18.435*** 14.519*** 31.578***
[12.41] [12.62] [12.77] [3.36] [4.10]
Industry Yes Yes Yes Yes Yes
Year Yes Yes Yes Yes Yes
Pseudo R2 0.29 0.31 0.32 0.46 0.48
Observations 1429 1429 1429 342 155
11
(continued on next page)
Table 3 (continued)
12
Panel B: Categories of RPTs, political connections, and auditor choice
Operating RPTs RPT loan and guarantees Operating RPTs and RPT loans
(1) (2) (3) (4) (5) (6) (7) (8) (9)
Predicted Full sample Only PCON = 1 Only PCON = 1 Full sample Only PCON = 1 Only PCON = 1 Full sample Only PCON = 1 Only PCON = 1
sign sample sample sample sample sample sample
GCON = 1 MCON = 1 GCON = 1 MCON = 1 GCON = 1 MCON = 1
Variables Coefficient [z- Coefficient [z- Coefficient [z- Coefficient [z- Coefficient [z- Coefficient [z- Coefficient [z- Coefficient [z- Coefficient [z-
stat] stat] stat] stat] stat] stat] stat] stat] stat]
PCON 0.22 – – 0.338* – – 0.125 – –
[1.19] [1.92] [0.66]
OPRPT_RATIO + 0.47*** 0.185 0.348 – – – 0.430*** 0.333 0.380

[3.40] [0.75] [0.55] [3.08] [1.26] [0.59]
LOAN_RATIO – – 1.180** 3.160*** 0.882 1.058** 3.719*** 0.880
[2.42] [3.49] [0.84] [2.19] [4.07] [0.83]
PCON⁄OPRPT_RATIO + 0.517 – – – – – 0.56 – –
[1.16] [1.29]
PCON⁄LOAN_RATIO – – – 1.373* – – 1.550** – –
[1.73] [1.96]
Other control Yes Yes Yes Yes Yes Yes Yes Yes Yes
variables
Industry Yes Yes Yes Yes Yes Yes Yes Yes Yes
Year Yes Yes Yes Yes Yes Yes Yes Yes Yes
Pseudo R2 0.29 0.37 0.47 0.31 0.42 0.47 0.31 0.42 0.47
Observations 1429 342 155 1429 342 155 1429 342 155
Variable definitions are in Appendix C.

***
Represent statistical significance at the 1% level (two-tailed test)
**
Represent statistical significance at the 5% level (two-tailed test).
*
Represent statistical significance at the 10% level (two-tailed test).
coefficient on LOAN_RATIO is also significantly negative for GCON sub-sample (Column 5) (coefficient 3.16, z-statistic 3.49,
p < 0.01). Finally, Column (7) reports results when both categories of RPTs are included in the same regression models, since
firms are likely to conduct both operating RPTs as well as RPT loans. Again we find the coefficient on OPRPT_RATIO and
LOAN_RATIO to be positive and negative respectively. Importantly the coefficient on the interactive variable
PCON⁄LOAN_RATIO is negative and significant (coefficient 1.55, z-statistic 1.96, p < 0.05) but that on PCON⁄OPRPT_RATIO
is insignificant (coefficient 0.56, z-statistic 1.29). This again provides corroborative evidence that RPT loans conducted by
connected firms primarily drive connected firms’ decision to appoint non-Big 4 auditors.
5.3. Endogeneity tests
A firm’s decision to get politically connected is not random, and unobservable factors that affect this decision may also be
associated with the propensity to choose a certain type of auditor. Selection problem, in this case, arise because one observes
only the outcome of the choice, in this case, political connection (1/0) made but not the outcomes of choices not made. Selec-
tion bias, which is one form of endogeneity problem, can lead to inappropriate inferences about treatment effects (Tucker,
2010). ‘‘Selection bias due to observables” arises from a failure to control for differences researchers can observe, e.g., size,
growth, complexity, profitability. ‘‘Selection bias due to unobservable” arises because researchers use a small set of obser-
vations. In the extant literature Heckman two-stage error correction method has been the most popular and widely used
approach for controlling the latter bias, whereas the ‘Propensity score matching’ (PSM) has been used to control for bias
due to unobservable factors.
To perform the Heckman test (1979) we proceed as follows. First we model firms’ decisions to form political connections
using some observable firm characteristics based on prior research (Faccio, 2006, 2010; Boubakri et al., 2008;
Bunkanwanicha and Wiwattanakantang, 2009). Lennox et al. (2012) argue that it is important to impose exclusion restric-
tions in implementing the Heckman two-stage regression, even though the inverse Mills ratio (IMR) can be identified by its
nonlinear arguments. In other words, we need at least one variable in the first-stage model that affects auditor choice
through its effects on political connections only.
Following Kim and Zhang (2016) and Guedhami et al. (2014), we include Industry% of connected firms (%PCON_IND)
and firms’ location (HQ) respectively, as the exclusion variables. The variable%PCON_IND is related positively to the polit-
ical connections of each individual firm within the industry. However, we have no a priori reason to believe that
industry-level political affiliation has a direct impact on auditor choice through channels other than political connec-
tions. Prior research suggests that firms located in capital cities are more likely to form political connections
(Agrawal and Knoeber, 2001), but are unlikely to be associated with auditor choice. Our first-stage probit model takes
the following form:
PCON ¼ c1 %PCON IND þ c2 HQ þ c3 OWNCON þ c4 FOWN þ c5 SIZE þ c6 LEV þ c7 GROWTH þ c8 SEGMENT þ c9 ROA
þ Industry FE þ YearFE þ e ð5Þ
%PCON_IND is the percentage of politically connected firms in a firm’s industry group. HQ is a dummy variable coded 1 if
the firm is headquartered in the capital city and zero otherwise. Other variables are defined as before. Table 4, Column (1)
reports the first stage estimation model. As predicted, the coefficients on %PCON_IND and on HQ, are significantly positive
[(coefficients 1.311 (p < 0.1) and 0.219 (p < 0.05)]. We calculate IMR from the first stage probit model, and include it as an
additional independent variable in the second stage regression model. We continue to find results that are consistent with
results in Panel B in Table 3. The coefficient on the interactive variable PCON⁄LOAN_RATIO is negative and marginally signif-
icant (coefficient 1.51, z-statistic 1.89, p < 0.10), while that on PCON⁄OPRPT_RATIO is insignificant. We find the coefficient
on IMR to be insignificant suggesting that self-selection does not confound our results.
5.3.1. Propensity-matched (PSM) technique

PSM technique is utilized to mitigate selection problem arising from observables. Matching on firm characteristics
(covariates) is ideal when the number of characteristics over which the treated and control groups differ is limited.
Rosenbaum and Rubin (1983) propose matching by a function of covariates: the probability of an individual being selected
into the program (treatment group). This matching method is referred to as ‘‘propensity score matching” (PSM). We use
nearest neighbor (NN), and kernel technique to perform the PSM model. The NN procedure with replacement picks a sin-
gle control firm according to the closest propensity score. Kernel matching uses the entire sample of control firms as
matches, where each unit is weighted in proportion to its closeness to the treated observation (Rosenbaum and Rubin,
1983, 1985).
Proper implementation of PSM requires both the treatment and control groups to be similar across a number of firm char-
acteristics excluding the main variable on which they are expected to differ, in our case, auditor. Therefore, we first docu-
ment the covariates matching, based on the calculated propensity score and (if necessary) remove the most dissimilar
matched pairs to achieve better control for potentially confounding factors (Armstrong et al., 2010, p. 240 and footnote
15). Results are reported in Panel A in Table 5. Covariate balance is achieved if both the treatment and control groups appear
similar along their observable dimensions, except for their choice of auditors. The p-values for the t-test indicate that the
Table 4
Endogeneity tests.
Heckman (1979) self-selection tests

PCON ¼ c1 %PCON IND þ c2 HQ þ c3 OWNCON þ c4 FOWN þ c5 SIZE þ c6 LEV þ c7 GROWTH þ c8 SEGMENT þ c9 ROA þ Industry FE þ YearFE þ e ð5Þ
AUDITOR ¼ c0 þ c1 PCON þ c2 OWNCON þ c3 FOWN þ c4 SIZE þ c5 LEV þ c6 GROWTH þ c7 INV þ c8 SEGMENT þ c9 FINANCE þ c10 ROA þ c11 AC þ c12 IC
þc13 jDACj þ c14 IMR þ Industry þ Year þ e ð1Þ
(1) (2) (3) (4) (5)
1st stage probit Expected sign for the PCON = 1 & PCON = 1 &
model 2nd stage variables GCON = 1 MCON = 1
DV = PCON
Variables Coefficient [z- Coefficient [z- Coefficient [z-stat] Coefficient [z-stat]
stat] stat]
PCON – 0.029 0.199 – –

[0.15] [1.07]
RPT_RATIO – 0.310⁄⁄ – – –
[2.38]
OPRPT_RATIO – + – 0.390⁄⁄⁄ 0.338 0.493
[2.71] [1.29] [0.81]
LOAN_RATIO – – 1.120⁄⁄⁄ 3.266⁄⁄⁄ 1.337
[2.75] [3.80] [1.29]
PCON⁄RPT_RATIO – 1.146⁄⁄⁄ – – –
[3.97]
PCON⁄OPRPT_RATIO – + – 0.535 – –
[1.20]
PCON⁄LOAN_RATIO – – 1.510⁄ – –
[1.89]
HQ 0.219⁄⁄ NA – – – –
[2.16]
%PCON_IND 1.311⁄ NA – – – –
[1.76]
OWNCON 0.092 + 2.715⁄⁄⁄ 2.436⁄⁄⁄ 3.910⁄⁄⁄ 4.808⁄⁄⁄
[0.43] [8.24] [7.21] [4.35] [2.98]
FOWN 0.328⁄⁄ + 2.137⁄⁄⁄ 2.145⁄⁄⁄ 3.944⁄⁄⁄ 4.905⁄⁄⁄
[2.37] [5.04] [5.37] [5.69] [3.95]
SIZE 0.478⁄⁄⁄ + 0.866⁄⁄ 0.576 0.664⁄⁄⁄ 0.792
[14.49] [2.04] [1.41] [2.59] [1.34]
LEV 0.310 1.654⁄⁄ 0.954 1.647 7.587⁄⁄
[1.00] [2.25] [1.51] [0.86] [2.48]
GROWTH 0.032⁄⁄⁄ + 0.042 0.031 0.037 0.066
[3.00] [1.37] [1.04] [1.11] [0.46]
INV – + 1.076⁄⁄ 1.075⁄⁄ 3.384⁄ 2.623
[2.09] [2.03] [1.87] [1.41]
⁄⁄⁄
SEGMENT 0.172 + 0.069 0.181 0.269 0.115
[2.22] [0.34] [0.89] [0.72] [0.21]
FINANCE – ? 0.382 0.304 0.277 0.137
[1.60] [1.38] [0.57] [0.17]
ROA 0.515 + 1.691 1.683⁄ 11.204⁄⁄⁄ 0.344
[1.35] [1.62] [1.92] [3.44] [0.09]
AC – + 0.665⁄⁄⁄ 0.694⁄⁄⁄ 0.139 4.633⁄⁄⁄
[3.39] [3.33] [0.20] [4.54]
IC – + 0.188 0.023 0.728 1.407
[0.73] [0.08] [1.24] [1.22]
|DAC| – ? 0.043 0.037 0.723 8.041⁄
[0.05] [0.05] [0.26] [1.68]
IMR – ? 0.344 1.514 0.886 4.662
[0.16] [0.72] [1.02] [1.48]
⁄⁄⁄
Constant 9.742 17.425⁄ 17.63⁄⁄⁄ 13.587⁄⁄ 15.837
[11.90] [1.75] [8.75] [2.34] [1.15]
Industry Yes Yes Yes Yes Yes
Year Yes Yes Yes Yes Yes
Pseudo R2 0.26 0.31 0.30 0.44 0.44
Observations 1429 1429 1429 342 155
matching algorithm was successful in achieving balance for most covariates. In particular, 8 of the 10 t-tests are not statis-
tically significant in both the NN and Kernel techniques.
Panel B, Table 5 shows the PSM regression results using the NN approach in Columns (1) to (4) and Kernel technique in
Columns (5) to (8). We find results that are consistent with the main results: (i) politically connected firms choose non-Big 4
Table 5
Propensity-matched technique.
Panel A: Propensity-matched variables

NN method Kernel method
Variable Treated Control t-stat p-value Treated Control t-stat p-value
OWNCON 0.51 0.52 0.75 0.451 0.51 0.52 0.61 0.54
FOWN 0.23 0.29 3.15 0.002 0.23 0.28 2.49 0.013
SIZE 20.04 19.97 0.83 0.407 19.91 19.90 0.15 0.883
LEV 0.16 0.17 1.43 0.152 0.15 0.16 1.17 0.243
INV 0.14 0.14 0.16 0.874 0.15 0.14 0.07 0.941
SEGMENT 0.99 0.93 1.91 0.057 0.98 0.93 1.66 0.097
FINANCE 0.14 0.13 0.83 0.408 0.14 0.13 0.46 0.646
ROA 0.08 0.09 1.62 0.115 0.08 0.09 1.62 0.115
AC 0.89 0.89 0.08 0.939 0.89 0.89 0.05 0.963
|DAC| 0.09 0.10 0.74 0.458 0.09 0.10 0.34 0.737
PANEL B: Regression results
(1) (2) (3) (4) (5) (6) (7) (8)
NN NN NN NN Kernel Kernel Kernel Kernel
Variables Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient Coefficient
[z-stat] [z-stat] [z-stat] [z-stat] [z-stat] [z-stat] [z-stat] [z-stat]
PCON 0.613⁄⁄ – 0.401⁄ 0.569⁄⁄⁄ 0.765⁄⁄⁄ 0.360⁄ 0.679⁄⁄⁄

[3.54] [1.91] [2.79] [4.74] [1.77] [3.56]
GCON – 0.724⁄⁄⁄ – – – 1.203⁄⁄⁄ – –
[3.62] [6.22]
MCON – 0.967⁄⁄⁄ – – – 0.647⁄⁄⁄ – –
[4.20] [2.83]
SCON + – 2.696⁄⁄⁄ – – – 2.247⁄⁄⁄ – –
[3.85] [3.62]
⁄⁄⁄ ⁄⁄
RPT_RATIO – – 0.664 – – – 0.554 –
[2.81] [2.48]
OPRPT_RATIO + – – – 0.237 – – – 0.225
[1.00] [1.06]
LOAN_RATIO – – – 1.667⁄⁄⁄ – – – 1.276⁄⁄
[2.92] [2.38]
⁄⁄
PCON⁄RPT_RATIO – – 0.723 – – – 1.153⁄⁄⁄ –
[2.11] [3.36]
PCON⁄OPRPT_RATIO + – – – 0.016 – – – 0.012
[0.06] [0.04]
PCON⁄LOAN_RATIO – – – 2.047⁄ – – – 1.590⁄
[1.80] [1.76]
OWNCON + 2.980⁄⁄⁄ 3.332⁄⁄⁄ 3.753⁄⁄⁄ 3.429⁄⁄⁄ 2.192⁄⁄⁄ 2.331⁄⁄⁄ 2.872⁄⁄⁄ 2.508⁄⁄⁄
[6.96] [7.51] [8.20] [7.60] [5.56] [5.73] [6.83] [6.06]
FOWN + 4.040⁄⁄⁄ 4.356⁄⁄⁄ 4.230⁄⁄⁄ 4.181⁄⁄⁄ 2.141⁄⁄⁄ 2.124⁄⁄⁄ 2.221⁄⁄⁄ 2.171⁄⁄⁄
[12.35] [12.69] [12.50] [12.38] [6.69] [6.44] [6.56] [6.58]
SIZE + 0.855⁄⁄⁄ 0.840⁄⁄⁄ 0.973⁄⁄⁄ 0.882⁄⁄⁄ 0.748⁄⁄⁄ 0.783⁄⁄⁄ 0.906⁄⁄⁄ 0.793⁄⁄⁄
[9.97] [9.50] [10.10] [9.58] [8.90] [8.81] [9.47] [8.84]
LEV 1.317⁄⁄ 1.106 1.355⁄ 0.900 0.576 0.638 0.824 0.497
[1.97] [1.61] [1.95] [1.30] [0.89] [0.96] [1.22] [0.74]
GROWTH + 0.029⁄ 0.032⁄ 0.043⁄⁄ 0.040⁄⁄ 0.001 0.003 0.012 0.006
[1.72] [1.91] [2.43] [2.30] [0.04] [0.19] [0.66] [0.35]
INV + 0.326 0.889 0.771 0.910 1.044⁄ 1.452⁄⁄ 0.994 1.170⁄
[0.56] [1.44] [1.25] [1.44] [1.69] [2.29] [1.51] [1.78]
SEGMENT + 0.386⁄⁄ 0.172 0.365⁄⁄ 0.413⁄⁄ 0.145 0.206 0.158 0.072
[2.33] [0.99] [2.13] [2.42] [0.89] [1.22] [0.92] [0.43]
FINANCE ? 0.261 0.193 0.372 0.335 0.326 0.295 0.444⁄ 0.321
[1.07] [0.77] [1.43] [1.29] [1.34] [1.17] [1.72] [1.26]
ROA + 5.829⁄⁄⁄ 5.699⁄⁄⁄ 5.266⁄⁄⁄ 5.446⁄⁄⁄ 4.808⁄⁄⁄ 4.667⁄⁄⁄ 4.602⁄⁄⁄ 4.312⁄⁄⁄
[6.34] [6.24] [5.46] [5.67] [5.22] [4.93] [4.81] [4.55]
AC + 1.062⁄⁄⁄ 1.161⁄⁄⁄ 0.877⁄⁄⁄ 0.970⁄⁄⁄ 0.927⁄⁄⁄ 1.129⁄⁄⁄ 0.765⁄⁄⁄ 0.875⁄⁄⁄
[3.87] [4.06] [3.06] [3.44] [3.27] [3.80] [2.61] [3.01]
IC + 0.357 0.745⁄⁄ 0.514 0.379 0.142 0.067 0.016 0.146
[1.09] [2.12] [1.45] [1.09] [0.44] [0.20] [0.05] [0.43]
|DAC| ? 0.717 0.049 0.711 0.689 1.201 1.391 1.407 1.367
[0.78] [0.05] [0.75] [0.71] [1.26] [1.39] [1.42] [1.39]
Constant 17.826⁄⁄⁄ 18.295⁄⁄⁄ 20.159⁄⁄⁄ 18.379⁄⁄⁄ 15.018⁄⁄⁄ 16.078⁄⁄⁄ 17.872⁄⁄⁄ 15.473⁄⁄⁄
[9.26] [9.15] [9.54] [8.99] [8.36] [8.44] [8.87] [8.11]
Industry Yes Yes Yes Yes Yes Yes Yes Yes
Year Yes Yes Yes Yes Yes Yes Yes Yes
(continued on next page)

Table 5 (continued)
PANEL B: Regression results

(1) (2) (3) (4) (5) (6) (7) (8)
NN NN NN NN Kernel Kernel Kernel Kernel
Pseudo R2 0.43 0.45 0.45 0.44 0.29 0.33 0.34 0.31
Observations 1150 1150 1144 1144 1398 1398 1391 1391
auditors, (ii) firms with government and military connection choose non Big 4 auditor whereas firms with Suharto connec-
tion choose Big 4 auditors, (iii) connected firms with RPTs choose non Big 4 auditors which is more pronounced for RP loans.
Overall, our PSM analysis provides robust evidence about the association between political connection and auditor choice
and the mediating role of firm-level RPTs.
6. Conclusions
This study investigates the association between political connections and auditor choice in Indonesia. We also investigate
whether firm-level RPTs moderate the association between the two. Our study is motivated by conflicting views on whether
politically connected firms appoint reputable auditors. We find that politically connected firms in Indonesia tend to appoint
less reputable auditors. Connected firms have been found to engage in tunneling and rent-seeking activities in order to
establish and maintain their political connections. Given the value of political connections (e.g. Fisman, 2001), politically
connected firms may manipulate accounting numbers to conceal their true economic performance and, hence, prefer less
reputable auditors. We also find that connected firms’ preference for appointing less reputable auditors is more pronounced
when these firms engage in significant RPTs.
The findings from this study might be generalized to other emerging countries, which have characteristics similar to those
of Indonesia, where political connections hold a significant role, and politically connected firms conduct considerable RPTs.
The findings of the research might benefit prospective investors and minority shareholders, who have limited information
and knowledge of the true economic incentives for RPTs conducted by listed firms that have political connections. The
minority shareholders should be aware that RPTs might be used opportunistically by controlling shareholders of politically
connected firms to carry out tunneling that is followed by subsequent earnings management. In addition, with regard to
auditor choice, the findings of this study might benefit prospective investors and minority shareholders regarding the ratio-
nale as to why politically connected firms appoint non-Big 4 auditors. The research findings show that the detrimental effect
to minority interests of RPT loans conducted by politically connected firms justifies the regulatory restrictions on RPTs and
the disclosure requirements imposed on listed firms.
Appendix A. Audit environment in Indonesia
The Indonesian Institute of Accountants9 has responsibility for establishing accounting standards, auditing standards,
and ethics codes for accountants. The International aid agencies such as the World Bank and IMF and foreign investors
recommend that the government both enforce tougher disclosure rules and harmonize accounting practices in Indonesia.
In response to those requests, following the enactment of the Sarbanes-Oxley Act in July 2002, the Indonesian government
issued the Minister of Finance Regulation Number 423/KMK.06/2002 dated 30 September 2002 regarding Public Accountant
Services.10
Basioudis and Fifi (2004) claim that accounting standards and corporate governance regulations in Indonesia have been
influenced significantly by the developed western countries and, sometimes, go even further than those of western regula-
tions. They point to auditor rotation as stipulated in Regulation number 423/KMK.06/2002, in which audit firms are not
allowed to audit an entity for five consecutive years, and audit partners can audit an entity for only three consecutive years.11
Further, they argue that the requirements for audit rotation in Indonesia are more burdensome compared to those of other juris-
dictions in the world, as even the Sarbanes-Oxley Act did not require rotation of audit firms at all, or rotation of audit partners
until after five years.
All financial statements were to be prepared in accordance with the Indonesian accounting standards which are mostly
derived from the International and US accounting standards. Further, financial statements were to be audited according to
9
Ikatan Akuntan Indonesia (IAI - the Indonesian Institute of Accountants) is the organization officially recognized by the government. IAI was established on
23 December 1957.
10
The amendment of the previous Minister of Finance Regulation number 43/KMK.017/1997. This regulation was further revised by the issuance of the
Minister of Finance Regulation number 17/PMK.01/2008, dated 05 February 2008, and was upgraded to the law by the enactment of Law number 5, year 2011,
regarding Public accountants on 03 May 2011.
11
The rotation of auditors was revised by the issuance of the Minister of Finance Regulation number 17/PMK.01/2008 by which rotation of audit firms is now
required after six consecutive years (one year longer than the previous regulation).
the Indonesian auditing standards, which are significantly similar to the generally accepted auditing standards in the US. In
providing an opinion on the fairness of financial statements, auditors refer primarily to the Indonesian Accounting Standards,
followed by relevant domestic rules and regulations, and International Accounting Standards for matters that are not cov-
ered by the abovementioned regulations ((Basioudis and Fifi, 2004).
Wulandari and Rahman (2005) point out that even though the Indonesian accounting standards are substantially sim-
ilar to the International Accounting Standards, some Indonesian companies still have low quality financial reports, owing
to the lack of effective and efficient enforcement mechanisms. Further, they argue that auditors might have been respon-
sible for playing a role in the Asian financial crisis, as many of listed companies were granted unqualified opinions prior to
the crisis, but went bankrupt during the financial crisis. They claim that auditors had little understanding and awareness
of accounting or auditing standards. In addition, auditors’ opinions were sometimes based on the work of other profes-
sionals, such as appraisers, when they took for granted the value of certain assets belonging to the auditee (Wulandari
and Rahman, 2005), and auditors rely heavily on management representations without conducting further investigation
to gather corroborative evidence in supporting their audit opinions (Rahman, 2000 as cited in Wulandari and Rahman,
2005).
Empirical evidence on whether audit quality is valuable in Indonesia is not available. We used our data to provide some
evidence on this important issue. Since extant research has established that financial reporting quality, as proxied by less
earnings management, is better for firms audited by Big 4 and industry specialist audit firms, we regress abnormal accruals
on auditor identity and other control variables. Untabulated results show the coefficient on AUDITOR, is significantly negative
(coefficient 0.023, t-stat 2.41), suggesting that Big 4 auditors in Indonesia constrain earnings management. This result
may provide a perspective on connected firms’ reluctance to appoint Big 4 audit firms, i.e., to conceal their tunneling and
rent-seeking activities. We also regress audit opinion (OPINION) (a dummy variable coded 1 if clients receive a qualified audit
opinion, and 0 otherwise) on the AUDITOR indicator variable and other relevant determinants of OPINION. Untabulated
results show the coefficient on AUDITOR is negative and significant (coefficient 0.42, z-statistic 1.85, p < 0.10). However,
this effect is pronounced for the PCON = 1 group only (coefficient 0.76, z-statistic 1.75, p < 0.10). This suggests that con-
nected firms tend to get more qualified opinions compared to their non-connected counterparts and this, again, might
explain connected firms’ preference for non-Big 4 auditors.
Appendix B. Development of RPT regulations in Indonesia
Bapepam-LK enacted Regulation Number VIII.G.7, which requires extensive disclosure on RPTs in the notes to financial
statements. Regulation Number VIII.G.7 under the Decree of the Head of the Indonesian Capital Market Supervisory Agency
number KEP-06/PM/2000 dated 13 March 2000 requires issuers and listed firms to disclose the breakdown and the total
amount of RPTs under each category of assets, liabilities, sales, and purchases (expenses), along with their percentage to total
assets, total liabilities, total sales, and total purchases (expenses); nature, types and elements of RPTs; pricing policies and
terms of RPTs and a statement that RPTs are similar to those undertaken with third parties; reasons and basis for the recog-
nition of doubtful accounts for RPT receivables. If the amount of a RPT for each category exceeds IDR 1,000,000,000 (US$
101,96812) or if non-core business RPTs are conducted, separate disclosures and explanations are needed. From 25 June
2012, a new RPT regulation was enacted, i.e., Regulation Number VIII.G.7, under a Decree of the Head of the Indonesian Capital
Market Supervisory Agency number KEP-347/BL/2012, in which the disclosure requirements are quite similar, but RPTs are now
classified based on three categories of parties conducting the RPT, namely, family members, related parties, and government
affiliated.
In addition, Regulation Number IX.E.1 under the Decree of the Head of the Indonesian Capital Market Supervisory
Agency, number KEP-412/BL/2009, dated 25 November 2009, requires issuers and listed firms to announce to the public
any RPT, and report the evidence of that announcement to the Indonesian Capital Market Supervisory Agency no later than
two working days after the transactions are undertaken, except for RPTs having a value less than 0.5% of the firm’s paid
capital and less than IDR 5,000,000,000 (US$509,840), in which case no public announcement is required, but such an
RPT shall be reported to the Indonesian Capital Market Supervisory Agency. In addition, if RPTs are identified as conflict
of interest transactions13, they need to get approval from the independent shareholders, except for RPTs having a low value
as specified above.
Recent RPT cases in the United Kingdom involved an Indonesian listed firm. Asia Resource Mineral Plc (formerly Bumi
Plc) suffered fines in the amount of £4,651,200 from the Financial Conduct Authority because its subsidiary, namely, PT
Berau Coal Energy Tbk (listed in Indonesian Stock Exchange) conducted an RPT which breached UKLA Listing Rules
(Authority, 2015). Those two companies are controlled by Bakrie Group, and are classified as politically connected firms
in our study.
12
(1US$ = IDR 9807 in 2013 (DataStream)
13
According to Bapepam-LK rule IX.E.I, conflict of interest takes place in the transaction when the economic interest of the firm differs from the personal
interest of directors, commissioners, or controlling shareholders, such that the transaction might ruin the firm.
Appendix C. Variable definitions
Variables Definitions
AUDITOR Dummy variable, 1 for the firm audited by Big 4 auditors, 0 for otherwise
PCON Dummy variable, 1 for politically connected firms, 0 for otherwise
GCON Dummy variable, 1 for government connected firms, 0 for otherwise
MCON Dummy variable, 1 for military connected firms, 0 for otherwise
SCON Dummy variable, 1 for Suharto connected firms, 0 for otherwise
RPT_RATIO Gross RPT divided by total assets of the firm
OPRPT_RATIO Gross operating RPTs divided by total assets of the firm.
LOAN_RATIO Gross RPTLOAN divided by total assets of the firm
OWNCON Total percentage of shares owned by the five largest shareholders
FOWN Total percentage of shares owned by foreign institutional investors
SIZE Natural logarithm of total assets
LEV Total long term debt divided by total assets
GROWTH Market value of equity divided by book value of equity
INV The ratio of inventory to total assets
SEGMENT Natural logarithm of number of business segments
FINANCE A dummy variable that takes the value of one if the sum of new long-term debt plus new equity exceeds
20% of total assets
ROA Return on assets (earnings before extraordinary items plus discontinued operations for the preceding
year divided by total assets for the same year)
AC Dummy variable, 1 for the firm having audit committee, 0 for otherwise
IC Dummy variable, 1 for the firm having independent commissioners, 0 for otherwise
ACC Total accruals derived from earnings before extraordinary items and discontinued operations minus
operating cash flows.
|DAC| Absolute discretionary accruals calculated with the Modified Jones model (1995). To estimate DAC we
use the cross-sectional modified Jones model, controlling for firm performance (Kothari et al., 2005). We
estimate the following model for all firms in the same industry (using economic sector code):
ACC t =TAt1 ¼ c0 ð1=TAt1 Þ þ c1 ½ðDSALESt DDEBTORSt Þ=TAt1 þ c2 ðPPEt =TAt1 Þ þ c3 ðROAÞ þ et ð2Þ
The coefficient estimates from Eq. (2) are used to predict non-discretionary component of total accruals
(NDAC) for our sample firms. Thus, discretionary accruals is the residual from Eq. (2), i.e. DAC = ACC-NDAC
DSALES Change in sales from year t-1 to year t
DDEBTORS Change in accounts receivable from year t-1 to year t
PPE Gross property, plant, and equipment
ROA Return on assets (earnings before extraordinary items plus discontinued operations for the preceding
year divided by total assets for the same year)
References
Abbott, L.J., Parker, S., Peters, G.F., Raghunandan, K., 2003. The association between audit committee characteristics and audit fees. Audit. J. Pract.Theory 22,
17–32.
Agrawal, A., Knoeber, C.R., 2001. Do some outside directors play a political role? J. Law Econ. 44, 179–198.
Armstrong, C.S., Jagolinzer, A.D., Larcker, D.F., 2010. Chief Executive Officer equity incentives and accounting irregularities. J. Account. Res. 48, 225–271.
Basioudis, I.G., Fifi, F., 2004. The market for professional services in Indonesia. Int. J. Audit. 8, 153–164.
Berkman, H., Cole, R.A., Fu, L.J., 2009. Expropriation through loan guarantees to related parties: evidence from China. J. Bank. Fin. 33, 141–156.
Bhakti, I.N., Yanuarti, S., Nurhasim, M., 2009. Military politics, etnnicity and conflict in Indonesia. Centre for Research on Inequality, Human Security and
Ethnicity – Working Paper, 62, pp. 1–36.
Bona-Sánchez, C., Pérez-Alemán, J., Santana-Martín, D.J., 2014. Politically connected firms and earnings informativeness in the controlling versus minority
shareholders context: European evidence. Corp. Gov. 22, 330–346.
Boubakri, N., Cosset, J.C., Saffar, W., 2008. Political connections of newly privatized firms. J. Corp. Fin. 14, 654–673.
Boubakri, N., Cosset, J.C., Saffar, W., 2012a. The impact of political connections on firms’ operating performance and financing decisions. J. Fin. Res. 35, 397–
423.
Boubakri, N., Guedhami, O., Mishra, D., Saffar, W., 2012b. Political connections and the cost of equity capital. J. Corp. Fin. 18, 541–559.
Brown, R.A., 2006. The Rise of the Corporate Economy in Southeast Asia. Oxon, United Kingdom, Routledge.
Bunkanwanicha, P., Wiwattanakantang, Y., 2009. Big business owners in politics. Rev. Fin. Stud. 22, 2133–2168.
Bushman, R.M., Piotroski, J.D., Smith, A.J., 2004. What determines corporate transparency? J. Account. Res. 42, 207–252.
Carcello, J.V., Hermanson, D.R., Neal, T.L., Riley, R.A., 2002. Board characteristics and audit fees. Contem. Account. Res. 19, 365–384.
Chaney, P.K., Faccio, M., Parsley, D., 2011. The quality of accounting information in politically connected firms. J. Account. Econ. 51, 58–76.
Chen, C.L., Hsu, C.S., Kung, F.H., 2015. Political connections, managerial incentives and auditor choice: evidence from China. Pacific Account. Rev. 27, 441–
465.
Chen, J.J., Chen, P., Xiao, X., 2011. Related Party transactions as a source of earning management. Appl. Fin. Econ. 21, 165–181.
Chen, S., Wang, K., Li, X., 2012. Product market competition, ultimate controlling structure and related party transactions. China J. Account. Res. 5, 293–306.
Cheung, Y.L., Jing, L., Lu, T., Rau, P.R., Stouraitis, A., 2009. Tunneling and propping up: an analysis of related party transactions by Chinese listed companies.
Pacific-Basin Fin. J. 17, 372–393.
Child, J., 1994. Management in China During the Age of Reform. Cambridge University Press, Cambridge.
Claessens, S., Djankov, S., Lang, L.H.P., 2000. The separation of ownership and control in East Asian corporations. J. Fin. Econ. 58, 81–112.
Davis, L.R., Ricchiute, D.N., Trompeter, G., 1993. Audit effort, audit fees, and the provision of nonaudit services to audit clients. Account. Rev. 68, 135–150.
DeFond, M.L., Wong, T.J., Li, S., 2000. The impact of improved auditor independence on audit market concentration in China. J. Account. Econ. 28, 269–305.
DeFond, M.L., Zhang, J., 2014. A review of archival auditing research. J. Account. Econ. 58, 275–326.
Dinç, I.S., 2005. Politicians and banks: political influences on government-owned banks in emerging markets. J. Fin. Econ. 77, 453–479.
Dyck, A., Zingales, L., 2004. Private benefits of control: an international comparison. J. Fin. 59.
Faccio, M., 2006. Politically connected firms. Am. Econ. Rev. 96, 369–386.
Faccio, M., 2010. Differences between politically connected and nonconnected firms: a cross-country analysis. Fin. Mgt. Autumn, 905–927.
Faccio, M., Masulis, R.W., Mcconnell, J.J., 2006. Political connections and corporate bailouts. J. Fin. 61, 2597–2635.
Fan, J.P.H., Wong, T.J., 2005. Do external auditors perform a corporate governance role in emerging market? Evidence from East Asia. J. Account. Res. 43, 35–
72.
Financial Conduct Authority, 2015. Asia Resource Minerals plc (formerly Bumi plc) fined £4.65m for breaching the Listing Rules. Financial Conduct
Authority. Available: <https://www.fca.org.uk/news/asia-resource-minerals-plc-formerly-bumi-plc-fined-4-65m-for-breaching-the-listing-rules#>
(Accessed 19 October 2015).
Fisman, R., 2001. Estimating the value of political connections. Am. Econ. Rev. 91, 1095–1102.
Frye, T., Shleifer, A., 1997. The invisible hand and the grabbing hand. Am. Econ. Rev. 87, 354–358.
Fukuoka, Y., 2013. Indonesia’s ’democratic transition’ revisited: a clientelist model of political transition. Democratization 20, 991–1013.
Goldman, E., Rocholl, J., So, J., 2009. Do politically connected boards affect firm value? Rev. Fin. St. 22, 2331–2360.
Gow, I.D., Ormazabal, G., Taylor, D.J., 2010. Correcting for cross-sectional and time-series dependence in accounting research. Account. Rev. 85, 483–512.
Guedhami, O., Pittman, J.A., Saffar, W., 2009. Auditor choice in privatized firms: empirical evidence on the role of state and foreign owners. J. Account. Econ.
48, 151–171.
Guedhami, O., Pittman, J.A., Saffar, W., 2014. Auditor choice in politically connected firms. J. Account. Res. 52, 107–162.
Gul, F.A., 2006. Auditors’ response to political connections and cronyism in Malaysia. J. Account. Res. 44 (5), 931–963.
Habib, A., Haris, A., Jiang, H., forthcoming. Political connections and related party transactions: evidence from Indonesia. Int. J. Account. forthcoming.
He, X., Rui, O., Zheng, L., Zhu, H., 2014. Foreign ownership and auditor choice. J. Account. Public Pol. 33, 401–418.
Heckman, J.J., 1979. Sample-selection bias as a specification error. Econometrica 47, 153–162.
Hope, O.K., Kang, T., Thomas, W., Yoo, Y.K., 2008. Culture and auditor choice: a test of the secrecy hypothesis. J. Account. Public Pol. 27, 357–373.
Houston, J.F., Jiang, L., Lin, C., Ma, Y., 2014. Political connections and the cost of bank loans. J. Account. Res. 52, 193–243.
Humphrey, C., Loft, A., Woods, M., 2009. The global audit profession and the international financial architecture: understanding regulatory relationships at a
time of financial crisis. Account. Org. Soc. 34, 810–825.
IASB, 2009. Related Party Disclosures. International Accounting Standard 24. Available at: <http://www.iasplus.com/en/standards/ias/ias24>.
Jensen, M.C., Meckling, W.H., 1976. Theory of the firm: managerial behavior, agency costs and ownership structure. J. Fin. Econ. 3, 305–360.
Jian, M., Wong, T.J., 2010. Propping through related party transactions. Rev. Account. St. 15, 70–105.
Jiang, G., Lee, C.M.C., Yue, H., 2010. Tunnelling through intercorporate loans: the China experience. J. Fin. Econ. 98, 1–20.
Johnson, S., Mitton, T., 2003. Cronyism and capital controls: evidence from Malaysia. J. Fin. Econ. 67, 351–382.
Khanna, T., Palepu, K., 2000. Is group affiliation profitable in emerging markets? An analysis of diversified Indian business groups. J. Fin. 55, 867–891.
Khwaja, A.I., Mian, A., 2005. Do lenders favor politically connected firms? Rent provision in an emerging financial market. Quart. J. Econ. 120, 1371–1411.
Kim, C.F., Zhang, L., 2016. Corporate political connections and tax aggressiveness. Contem. Account. Res. 33, 78–114.
Knechel, R.W., Salterio, S., Ballou, B., 2007. Auditing: Assurance and Risk. South-Western College Pub.
Kothari, S.P., Leone, A.J., Wasley, C.E., 2005. Performance matched discretionary accrual measures. J. Account. Econ. 39, 163–197.
Kohlbeck, M., Mayhew, B.W., 2010. Valuation of firms that disclose related-party transactions. J. Account. Public Pol. 29, 115–137.
Lennox, C.S., Francis, J.R., Wang, Z., 2012. Selection models in accounting research. Account. Rev. 87 (2), 589–616.
Leuz, C., Oberholzer-Gee, F., 2006. Political relationships, global financing, and corporate transparency: evidence from Indonesia. J. Fin. Econ. 81, 411–439.
Li, H., Meng, L., Wang, Q., Zhou, L.A., 2008. Political connections, financing and firm performance: evidence from Chinese private firms. J. Dev. Econ. 87, 283–
299.
Ma, L., Ma, S., Tian, G., 2013. Political connections, founder-managers, and their impact on tunneling in China’s listed firms. Pacific-Basin Fin. J. 24, 312–339.
Mietzner, M., 2006. The Politics of Military Reform in Post-Suharto Indonesia: Elite Conflict, Nationalism, and Institutional Resistance. East-West Center
Washington, Washington, D.C..
Morck, R.K., Stangeland, D.A., Yeung, B., 2000. Inherited wealth, corporate control, and economic growth: the Canadian disease. In: Morck, R.K. (Ed.),
Concentrated Corporate Ownership. University of Chicago Press, Chicago.
Newman, D.P., Patterson, E.R., Smith, J.R., 2005. The role of auditing in investor protection. Account. Rev. 80, 289–313.
Petersen, M.A., 2009. Estimating standard errors in finance panel data sets: comparing approaches. Rev. Fin. Stud. 22, 435–480.
Piotroski, J.D., Wong, T.J., Zhang, T., 2015. Political incentives to suppress negative information: evidence from Chinese listed firms. J. Account. Res. 53, 405–
459.
Qian, M., Pan, H., Yeung, B., 2011. Expropriation of minority interest shareholders in politically connected firms. Working Paper – NUS Business School.
Rosenbaum, P.R., Rubin, D.B., 1983. The central role of the propensity score in observational studies for causal effects. Biometrika 70, 41–55.
Rosenbaum, P.R., Rubin, D.B., 1985. Constructing a control group using multivariate matched sampling methods that incorporate the propensity score. Am.
Stat. 39, 33–38.
Ryngaert, M.D., Thomas, S., 2012. Not all related-party transactions (RPTs) are the same: ex ante versus ex post RPTs. J. Account. Res. 50, 845–882.
Sebastian, L.C., Iisgindarsah, 2013. Taking stock of military reform in Indonesia. In: Rüland, J., Manea, M.G., Born, H. (Eds.), The Politics of Military Reform
Experiences from Indonesia and Nigeria. Verlag Berlin Heidelberg Springer.
Thomas, W.B., Herrmann, D.R., Inoue, T., 2004. Earnings management through affiliated transactions. J. Int. Account. Res. 3, 1–25.
Tucker, J.W., 2010. Selection bias and econometric remedies in accounting and finance research. J. Account. Lit. 29, 31–57.
Utama, C.A., Utama, S., 2013. Corporate governance, size and disclosure of related party transactions, and firm value: Indonesia evidence. Int. J. Disc. Gov., 1–
25
Utama, C.A., Utama, S., 2014. Determinants of disclosure level of related party transactions in Indonesia. Int. J. Disc. Gov. 11, 74–98.
Wang, Q., Wong, T.J., Xia, L., 2008. State ownership, the institutional environment, and auditor choice: evidence from China. J. Account. Econ. 46, 112–134.
Watts, R.L., Zimmerman, J.L., 1983. Agency problems, auditing, and the theory of the firm: some evidence. J. Law Econ. 26, 613–633.
Whisenant, S., Sankaraguruswamy, S., Raghunandan, K., 2003. Evidence on the joint determination of audit and non-audit fees. J. Account. Res. 41, 721–744.
Wulandari, E.R., Rahman, A.R., 2005. Political patronage, cross-holdings and corporate governance in Indonesia. In: Gul, F.A., Tsui, J.S.L. (Eds.), The
governance of East Asian corporations. Palgrave Macmillan, N.Y..
Ying, Q., Wang, L., 2013. Propping by controlling shareholders, wealth transfer and firm performance: Evidence from Chinese listed companies. China J.
Account. Res. 6, 133–147.

Journal of Contemporary Accounting & Economics: Ahsan Habib, Abdul Haris Muhammadi, Haiyan Jiang

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Journal of Contemporary Accounting & Economics: Ahsan Habib, Abdul Haris Muhammadi, Haiyan Jiang

Transféré par

Droits d'auteur :

Formats disponibles

Journal of Contemporary Accounting & Economics 13 (2017) 1–19

Contents lists available at ScienceDirect

Political connections, related party transactions, and auditor

2. Literature review and hypotheses development

3. Research design – empirical model

To test H1, we develop the following regression model:

4. Sample selection and descriptive statistics

Panel A: Sample selection procedure

Panel B: Descriptive statistics

Panel C: Industry distributions

5. Main test results

5.1. Political connections and auditor choice

5.2. Political connections, RPTs, and auditor choice

A. Habib et al. / Journal of Contemporary Accounting & Economics 13 (2017) 1–19

A. Habib et al. / Journal of Contemporary Accounting & Economics 13 (2017) 1–19

Variable definitions are in Appendix C.

5.3. Endogeneity tests

5.3.1. Propensity-matched (PSM) technique

Heckman (1979) self-selection tests

PCON – 0.029 0.199 – –

Panel A: Propensity-matched variables

PCON 0.613⁄⁄ – 0.401⁄ 0.569⁄⁄⁄ 0.765⁄⁄⁄ 0.360⁄ 0.679⁄⁄⁄

(continued on next page)

PANEL B: Regression results

Appendix A. Audit environment in Indonesia

Appendix B. Development of RPT regulations in Indonesia

Appendix C. Variable definitions

Vous aimerez peut-être aussi