CORPORATE POLITICAL CONNECTION AND AUDIT QUALITY

Indonesia offers interesting conditions to explore research on political connections because Indonesia is a country with an emerging market and a history of strong political connections in business. This study aims to empirically examine the effect of corporate political connections on audit quality and whether the effect of political connections on auditor selection is moderated by earnings management behavior. This study used a sample of all companies from all sectors listed on the Indonesia Stock Exchange (IDX) during the 2015-2017 period, except for companies from the financial industry sector, with a total of 1,213 firm-years. The analysis technique used was logistic regression. The results revealed that politically connected boards of commissioners and directors in a company had a positive and significant effect on audit quality, meaning that companies with political connections tended to select industry specialist auditors as the company’s external auditors. The results also revealed that earnings management did not moderate the effect of corporate political connections on audit quality. This study provides new insights on the use of political connections in the business environment, especially in the quality of audits in Indonesia.


INTRODUCTION
During the Joko Widodo and Jusuf Kalla (Jokowi-JK) administration, there were several names of people who were appointed as commissioners of State-Owned Enterprises (SOE). Some of these people were those who were connected as part of the national campaign team of Jokowi-JK or members of the supporting parties of Jokowi-JK administration (Asril 2015;Sudrajat 2017), for example, Cahaya Dwi Rembulan Sinaga, a politician from the Indonesian Democratic Party of Struggle (PDI-P) who was appointed as independent commissioner of PT Bank Mandiri; Jefry Wurangian (politician of the Nasdem Party) who was appointed as commissioner of PT Bank Rakyat Indonesia (BRI); Roy E. Maningkas (politician of PDI-P) who was appointed as commissioner of PT Krakatau Steel; Imas Aan Ubudiah (politician of National Awakening Party (PKB)) who was appointed as independent commissioner of PT Wijaya Karya; Dolfie Othniel Fredric P. (survey team coordinator of the national campaign team of Jokowi-JK) who was appointed as commissioner of Telkom Indonesia; Fadjroel Rachman (Coordinator of Relawan Dua Jari (Two-Finger Volunteers)) who was appointed as the chief commissioner of PT Adhi Karya, and many more. In Indonesia, a corruption scandal in a government project committed by Teuku Bagus Mokhamad Noor, top brass of PT Adhi Karya, Tbk, who was found guilty of offering bribes to a number of government officials and members of the House of Representatives (DPR) to enable the enterprise to ob-tain contracts for the construction of Education and Training Center and the National Sports School in Bukit Hambalang, illustrates that enterprise's top officials may abuse authority for their personal benefit or the benefit of others (Maharani 2014).
This phenomenon illustrates that the involvements of politicians in business, business people in politics, state civil officials in related political and business activities, and also the dependence of political parties on contributions from companies/businesses as a source of funding contribute to the significance of political connections in business (Narayanaswamy 2013). Executive officers who have political connections are usually considered powerful because they can take advantage of various privileges such as to strengthen their position and influence company profits (Khan et al. 2016).
The effect of political connections on companies has drawn a lot of attention from several previous studies, such as studies on the effect of political connections on access to finance (Claessens et al. 2008), firm value (Fisman 2001), tax rates (Adhikari et al. 2006), tax aggressiveness (Kim and Zhang 2016), loan costs (Houston et al. 2014), and earnings quality (Chi et al. 2016;Cho and Song 2017;Hashmi et al. 2018;Narayanaswamy 2013). Previous studies have shown that political connections may exist in both developed and developing countries, including Indonesia. Indonesia is a country with a developing market and a history of strong political connections in business (Harymawan and Nowland 2016). Companies that have political connections in Indonesia are very dependent on the benefits of their connections (Fisman 2001). Faccio (2006) stated that a company was said to have political connections if one of the company's executives (CEO, president, vice president, chairman, or secretary) or one of the largest shareholders (at least shareholder who controls 10% of the total shares with voting rights) ever or currently served as a parliament member, minister, or individual that was closely related to politicians or political parties. In addition to providing benefits to business continuity, corporate political connections can influence the decision to select an external auditor (Cheng et al. 2015;Guedhami et al. 2014;Habib et al. 2017;He et al. 2015;Liu et al. 2017) . The decision to select an independent external auditor as a monitoring role on behalf of the owner/shareholder becomes an important component of corporate governance because it can help improve the credibility of accounting information provided by management (Abdel-Khalik 2002;Lin and Liu 2009). The reason why companies select certain auditors is complex and tends to vary across companies (Knechel et al. 2008;Lin and Liu 2009).
There is a difference of opinion between hiring a qualified auditor (as an indication that the company is conducting effective audit monitoring and good corporate governance to reduce the cost of increasing company capital) and hiring a less qualified auditor (to maintain earnings management due to poor corporate governance and less transparent disclosures). Differences in opposing views related to companies with political connections in their decision to select an external auditor have been illustrated in several previous studies. From a positive point of view, Guedhami et al. (2014) revealed that companies with political connections would tend to appoint high-quality auditors (from "Big Four" accounting firms) as a binding mechanism to reduce agency prob-lems and suspicion of information asymmetry. In addition, companies with political connections were more likely to appoint high-quality auditors to show that people within these companies wanted to increase accounting transparency and convinced investors that they refrained from exploiting their connections to divert corporate resources. In agency theory, the appointment of auditors with high competence by companies with political connections reflects bonding costs to reduce residual agency costs.
From a negative point of view, Cheng et al. (2015) revealed that companies with political connections (CEOs who had connection to politics, local-owned enterprises, and state-owned enterprises) tended to employ poor-quality auditors (non-top-10 auditors). It indicated that the company was less concerned with monitoring and the role of audit information. Corporate managers with political connections who pursue their own political profits tend to appoint underqualified auditors (non-top-10 auditors) as a means of covering up their earnings management behavior. The company does not want to appoint industry specialist auditors to improve audit quality because the stock market does not always offer sufficient incentives. Habib et al. (2017) also supported the finding and stated that politically connected companies in Indonesia tended to appoint poor-quality (less-reputable/non-big-4) auditors. Companies with political connections tended to appoint non-big-4 auditors when these companies were involved in significant third-party transactions.
Some of these opposing views indicate that corporate political connections have influenced the decision to select qualified or less-qualified auditors. Specifically, politically connected business leaders have an opportunistic incentive to manage earnings in order to maintain reputation and reduce political costs (Hu et al. 2012) so they are more likely to appoint less-qualified auditors to cover these suspicious activities (Cheng et al. 2015) . Politically connected business leaders prefer to hire auditors who will be more willing to hide specific information and dubious company activities in order to protect their political interests (Guedhami et al. 2009). The demand for good-quality auditors will decrease when the incentive for politically connected company officials to manage earnings increases. Thus, earnings management may moderate the relationship between political connections and auditor selection decisions.
Based on the phenomena and previous studies, this study aims to empirically examine the effect of political connections of people in the company on audit quality and whether the effect of political connections on audit quality is moderated by earnings management behavior. This study develops a previous study (Cheng et al. 2015;Guedhami et al. 2014;Habib et al. 2017;He et al. 2015;Liu et al. 2017;Narayanaswamy 2013) with some motivation. The first motivation is the existence of opposing views from previous studies regarding the effect of corporate political connections on auditor selection decision, which becomes a research gap. The second motivation is that there have been several previous studies in Indonesia (Butje and Tjondro 2015;Hardianti 2014;Lestari and Putri 2017;Mulyani et al. 2016;Siregar and Widyawati 2016) that discussed a lot about the effect of political connections that lead to tax aggressiveness and tax avoidance, so now, this study tries to give an insight into the effect of corporate political connections on audit quality. Furthermore, the third motivation is previous research that discussed the effect of political connections on auditor selection (Cheng et al. 2015;Guedhami et al. 2014;Habib et al. 2017;He et al. 2015;Liu et al. 2017;Narayanaswamy 2013). The majority used proxy measures of Public Accounting Firms, which were assumed to be goodquality auditors and poor-quality auditors, so this study tries to use another measurement dimension of audit quality with the proxy of industrial specialist auditors and industrial non-specialist auditors. Industry specialist auditors play an important role in providing high-quality audit because specialist auditors must have practical experience gained from auditing in certain industries, in addition to specific knowledge (Habbash and Alghamdi 2017). The fourth motivation is that previous studies (Cheng et al. 2015;Guedhami et al. 2014;Habib et al. 2017;He et al. 2015; Narayanaswamy 2013) used political connection measures and used binary indicator variables to indicate whether a company had political connections, so variations in the strength of connections were less revealed. This study attempts to establish measurements of political connections used by Liu et al. (2017), which employed political connection score index with various position levels by adjusting to the context in Indonesia.
This study used a sample of all companies from all sectors listed on the Indonesia Stock Exchange (IDX) for the period 2015-2017. This period was chosen because this study wanted to reveal the political conditions in the Jokowi-JK administration based on the phenomenon of the members of campaign team and supporting parties of Jokowi-JK who served as state-owned enterprises commissioners after the 2014 Presidential Election. Indonesia offers interesting conditions to explore this research topic because political connections affect corporate value (Fisman 2001) and Indonesia is a country with a developing market and a history of strong political connections in business (Harymawan and Nowland 2016).
This study provided empirical evidence that during the 2015-2017 period, politically connected companies in Indonesia did not use their political connections to select auditors who could protect and cover companies' political interests. The results of this study might also provide an overview for investors and other institutions (private, public, or state-owned companies) regarding the reasons why politically connected companies appoint industry specialist auditors as their external auditors. Then, this study also provided other measurement alternatives to the variables of political connections and audit quality that can be used for further research. The analysis technique used in this study was logistic regression analysis.

Audit Quality (Industry Specialist and Non-Specialist Auditors)
Agency problems increase the need for third-party guarantees and audit quality (Alfraih 2017). Cited from Alfraih (2017) and Abbott and Parker (2000), Dopuch and Simunic (1980) argued that audit requirements were closely related to agent needs. The decision to select independent external auditor as a monitoring role on behalf of owner/shareholder becomes an important component of corporate governance because it can help improve the credibility of accounting information provided by management (Abdel-Khalik 2002; Lin and Liu 2009). However, external audits only provide valuable information about companies' financial statements when they are of sufficient quality (Quick et al. 2013). Chang et al. (2009) assumed that all external auditors might not offer the same quality of service.
The reason why companies select certain auditors is complex and tends to vary across companies (Knechel et al. 2008;Lin and Liu 2009). There is a trade-off between hiring qualified auditors (as an indication that a company is conducting effective audit monitoring and good corporate governance to reduce the cost of increasing company capital) and hiring less-qualified auditors (to maintain earnings management as a result of poor corporate governance and less transparent disclosures).
The auditor has the duty to provide certainty to the credibility of the information presented in financial statements. The knowledge of an auditor is not only limited to auditing and accounting, but an auditor must also understand their client's business industry. Balsam et al. (2003) argued that industry specialist auditors offered a higher level of assurance than non-specialist auditors because of their specialized knowledge of the industry and accounting in the industry. Therefore, the appointement of auditors who specialize in industry will help reduce earnings management. Auditors who specialize in their client's industry are expected to provide higher level of audit quality than non-specialist auditors (Abbott and Parker 2000). Industry specialist auditors play an important role in providing high-quality audit because specialist auditors must have practical experience gained from auditing in certain industries in addition to specialized knowledge (Habbash and Alghamdi 2017).
Analysis from Porter (1985) showed that public accounting firms had the motivation to provide special services that met the needs of clients by differentiating themselves from their competitors. Differentiation may occur on many dimensions of client characteristics and demands of related service. Dimensions of these characteristics include size, number of segments, industry affiliation, regulations, capital resources, and relations of management/audit personnel. Mayhew and Wilkins (2003) argued that the client's industry affiliation was an important dimension that public accounting firms could use to align themselves with certain client characteristics and service needs. The market share of public accounting firm with industry specialization reflects the efforts made by auditors to specialize in their client's industry.

Effects of Political Connections on Audit Quality
Based on agency theory proposed by Jensen and Meckling (1976), companies with political connections are closely related to the emergence of agency conflicts, tend to minimize the disclosure of key information (Rodríguez et al. 2007), and tend to issue poor quality financial statements (Al-Dhamari and Ku Ismail 2015; Chung et al. 2005). Companies with political connections have little incentive to report high-quality earnings compared to companies without political connections (Chaney et al. 2011;Hashmi et al. 2018;Narayanaswamy 2013). Cheng et al. (2015) revealed a result that there was a significant negative effect of corporate political connections on auditor selection decision. These results indicated that companies with political connections tended not to employ qualified auditors (non-top-10 auditors). It also indicated that a company lacked awareness of the role of monitoring and audit information. Corporate managers with political connections who pursue their own political profit tend to appoint under-qualified (non-top-10) auditors as a means of covering up their earnings management behavior. In addition, companies with political connections do not want to appoint industry specialist auditors to improve audit quality because the stock market does not always offer sufficient incentives. Habib et al. (2017) also revealed the results that corporate political connections had a significant negative effect on auditor selection decisions. This result indicated that companies with political connections in Indonesia tended to appoint under-qualified auditors (non-big-4 auditors). Companies with political connections generally manipulate quantitative data in financial statements to hide their actual economic performance, so they are more likely to select less-qualified auditors. Companies with political connections tend to appoint less-qualified auditors when the company is involved in significant third-party transactions. Companies with political connections avoid big-4 auditors because big-4 auditors are considered as high-quality auditors who are competent to detect corporate misconduct. In addition, this is also because the big-4 public accounting firms avoid risky clients to reduce litigation costs and maintain their reputation.
As such, it makes sense to hypothesize that companies with a board of commissioners or board of directors who have political connections are more likely to appoint nonspecialist auditors to hide information about the company's actual performance as contained in the financial statements in order to strengthen their political policies and objectives and to maintain earnings management due to poor corporate governance and lesstransparent disclosures. With this background, the hypothesis formulated is as follows:

H1: Corporate political connections
negatively affect audit quality. Liu et al. (2018) documented that rentseeking activities lead to higher earnings management (in the case of discretionary accruals), which in turn results in poorquality accounting information being revealed to the market. Rent-seeking can be evaluated within a theoretical economic framework of self-preservation (Karni and Schmeidler 1986). In this framework, rational executives that are involved in rentseeking activities in companies tend to take appropriate actions that lead to bad actions for self-preservation.

Earnings Management as a Moderating Role of The Effect of Corporate Political Connections on Audit Quality
Given the drive to protect oneself, there are two main reasons that link rentseeking with earnings management. First, Liu et al. (2018) argued that when rentseeking companies were involved in corruption, bribery, or reciprocal relationship to get assistance from government officials, company's executives tended to take action to preserve themselves. That was because rentseeking activities made company's executives nervous and damaged the morale of employees if the employees who were involved in it became known to the public or were caught committing acts of wrongdoing.
If rent-seeking company's executives are rational, they will take a conservative approach to conceal the benefits of their rent-seeking activities by revealing a little company profile to minimize public scrutiny. As such, executives will cover up reliable information to protect themselves from public scrutiny through earnings management. Earnings management is used to hide company-specific information, so it becomes an important indicator of accounting information quality (Liu et al. 2018). Scott (2003) stated that one of the motivations of managers to manage earnings was political motivation.
The second motivation was the risk of political gains related to self-preservation. After accumulating economic benefits from the government, rent-seeking companies can attract the attention of corrupt government officials. Corrupt government officials have an incentive to exploit the situation for their personal interests. That is, officials can be involved in political profits from rentseeking companies. To minimize this risk of political profits, rent-seeking companies have incentives to reduce corporate governance and hide their investment and cash holdings (Caprio et al. 2011;Smith 2016).
Rent-seeking theory is associated with companies with political connections and earnings management because rent-seeking involves economic activities that directly or indirectly benefit government officials (Liu et al. 2018). Companies with political connections are closely related to the emergence of agency conflicts, tend to minimize the disclosure of key information (Rodríguez et al. 2007), and tend to report poor quality financial statements (Al-dhamari and Ku Ismail 2015; Chung et al. 2005). Specifically, politically connected company's executives have an opportunistic incentive to manage earnings in order to maintain reputation and reduce political costs (Hu et al. 2012).
Previous studies have shown that there is a negative relationship between corporate political connections and earnings quality (Chaney et al. 2011;Hashmi et al. 2018;Narayanaswamy 2013). It means that companies with political connections have little incentive to report high-quality earnings compared to companies with no political connections. Politically connected companies in Indonesia avoid foreign financing that demands greater transparency and makes it difficult for them to gain personal benefits from political connections (Leuz and Oberholzer-Gee 2006). Therefore, the executives of politically connected companies tend to prefer to hire auditors who will be more willing to hide specific information and dubious company activities in order to protect their political interests (Guedhami et al. 2009). Thoughts that lead to managerial opportunism can encourage management to appoint lessqualified auditors to ensure their opportunistic practices remain hidden. Based on this argument, the hypothesis formulated is as follows: H2: Earnings management reinforces the negative effect of political connections on audit quality.

Samples and Data Sources
This study sampled all companies from all sectors listed on the Indonesia Stock Exchange (IDX) for the period 2015-2017, except for companies from the financial sector. Companies from the financial industry sector (SIC 8) were excluded from the criteria because of the different nature of their financial statements and accounting practices (Habbash and Alghamdi 2017;Harymawan and Nowland 2016). The research sampling technique was purposive sampling. The criteria for the company being sampled were: 1. Companies that were listed on the IDX during the 2015-2017 period 2. Companies that were not delisted during the study period 3. Companies that provided data needed in their annual and financial statements 4. Financial statements audited and ended on December 31 Based on the criteria, the total sample of companies obtained in this study was 1,213 firm-years.

Data Analysis Techniques and Research Models
The equation model used in this study are:

Annotation:
ACi,t = audit quality of company i in period t PCONi,t = corporate political connections of corporate i in period t DACC i,t = discretionary accruals (DACC) of company i in period t The first hypothesis was tested using Equation Model 1, while the second hypothesis was tested using Equation model 3. The analysis technique used was logistic regression analysis and moderated regression analysis (MRA), which were processed with the help of WARP PLS 6.0.

Measurement of Audit Quality
The dependent variable in this study was audit quality. In this study, audit quality variable was measured by using industrial specialist and non-specialist auditors. Previous studies discussing the effect of political connections on audit quality (Cheng et al. 2015;Guedhami et al. 2014;Habib et al. 2017;He et al. 2015;Liu et al. 2017; Narayanaswamy 2013) mostly used proxy measure of public accounting firm that was assumed to be qualified and less-qualified auditors. Thus, this study employed another measurement dimension of auditor selection with proxies of industry specialist and nonspecialist auditors. Auditors with certain industry specialization are expected to provide a higher level of audit quality than non-specialist auditors (Abbott and Parker 2000). Industry specialist auditors play an important role in providing high-quality audit because specialist auditors must have practical experience gained from auditing in certain industries in addition to special knowledge (Habbash and Alghamdi 2017).
Most of the previous studies (Alzoubi 2017;Balsam et al. 2003;Habbash and Alghamdi 2017;Mayhew and Wilkins 2003) showed that industry specialist auditors played an important role in providing highquality audit. Balsam et al. (2003) argued that industry specialist auditors offered a higher level of assurance than non-specialist auditors because of the auditor's specialized knowledge of the industry and accounting in the industry. Therefore, the appointment of industry-specialist auditors will help reduce earnings management. Auditors who specialize in the client's industry are expected to provide higher level of audit quality than non-specialist auditors (Abbott and Parker 2000).
In this study, the market share approach was used as a measurement for industry specialist auditors. It is because this approach illustrates that public accounting firms have a greater market share compared to their comepetitors in certain industries. Thus, public accounting firms with the largest market share have improved knowledge in certain industries and indicated important investments by developing the industry in certain audit technologies (Lowensohn et al. 2007;Neal and Riley 2004). Mayhew and Wilkins (2003) argued that public accounting firms with large market shares were able to widen industryspecific knowledge and expertise that lead to higher quality services than the quality service of public accounting firms with small market shares. A public accounting firm is considered a specialist in the industry when it has the largest market share based on cli-ent sales or the number of clients audited by the firm. In addition, public accounting firms with the largest market share can develop their knowledge in the industry and increase their understanding of sensitive issues related to the industry. A public accounting firm is categorized as an industry specialist if it controls a market share of 20% or more of the total companies in the industry (Mayhew and Wilkins 2003). The calculation of industry specialist auditors (AC) in this study only covered public companies. The percentage of industry specialist auditors (AC) is calculated as follows: Furthermore, the value of the industry specialization auditor was measured by a dummy variable that was coded as 1 if the public accounting firm was an industry specialist (controlled market share) and coded as 0 if the public accounting firm was not an industry specialist.

Measurement of Political Connection
Faccio (2006) stated that a company was said to have political connection if one of the company's executives (CEO, president, vice president, chairman, or secretary) or one of the largest shareholders (at least shareholder who controlled 10% of the total shares with voting rights) has ever served or currently serves as parliament member, minister, or individual closely related to politicians or political parties. In Indonesia, the Indonesian Financial Transaction Reports and Analysis Centre (INTRAC) regulates the categories of service users who have the potential to commit money laundering. In the regulation of the INTRAC (2015, p. 3), it is stated that Politically Exposed Person (abbreviated as PEP) is defined as: From this definition, individuals who can be categorized as PEP are, for example, the president and vice president of Indonesia; minister and equivalent position; governor and vice governor; regent or mayor; members of the People's Consultative Assembly that includes the House of Representatives and the Regional Representative Council; heads and directors of state-owned enterpises and local-owned enterprises, and heads of political parties, high-ranking military and police officers; high-ranking members of the judiciary; managers or members of political parties and other positions mentioned in this regulation, as well as immediate family members of PEP, including second-degree relatives. If the board of directors and the board of commissioners have one of the criteria above, then it can be said that the company has political connections.
In this study, the measurement criteria for defining companies that had political connections (PCON) was the approach that was proposed by Liu et al. (2017), which used a political connection score index with various ranks of positions in the bureaucracy. Measurement criteria with the approach of Liu et al. (2017) was used by adjusting to the context of regulation of the INTRAC in Indonesia. Steps in measuring companies that had political connections (PCON) were based on the approach that was proposed by Liu et al. (2017) with some modifications. The first modification was the identification of politically connected directors and com-missioners based on informations about personal career background that were collected manually from company annual reports via searches on Google.
The second modification was by giving a score to company's executives who had political connections in accordance with the highest classification of positions based on regulations in Indonesia. In the regulation of the INTRAC (2015), Politically Exposed Person or PEP is classified into 5 categories (see Table 1) so that this study added 5 scores based on the classification of the highest to the lowest positions. The third modification was by giving a score of 5 for the classification of the highest position and a score of 1 for the classification of the lowest position. A score of 0 was given if the company had no political connections. Highlevel positions have greater political power and greater access to political and economic resources. Thus, the highest to lowest position classification provided a measure of the level of political relations. If a company's executive currently holds (or has previously held) several positions in the government or politics mentioned in Table 1, then all scores from these positions were added up and averaged when connection scores were input. Meanwhile, the fourth scores were input. Meanwhile, the fourth modification was that after inputting the scores of each company's executive, the scores of each company were then added together.

Measurement of Profit Management
Earnings management variable was proxied using discretionary accruals from the modified Jones model. The modified Jones model was used because compared to other models, this model is considered as the best model in detecting earnings management as well as providing the most reliable results (Dechow et al. 1995). In this study, regression was conducted for each industry which was divided into 8. 1) Agriculture Sector (SIC 1) 2) Mining Sector (SIC 2) 3) Basic Industry and Chemicals (SIC 3) 4) Miscellaneous Industry Sector (SIC 4) 5) Consumer Goods Sector (SIC 5) 6) Property, Real Estate, and Building Construction Sector (SIC 6) 7) Infrastructure, Utilities, and Transportation (SIC 7) 8) Trade, Services, and Investment (SIC 9) After conducting a regression in each industry, the value of discretionary accruals (DACC) was made absolute.

Measurement of Control Variables
This study used control variables that referred to Alfraih (2017), Cheng et al. (2015), Guedhami et al. (2014), andHabib et al. (2017). Variables that were not included in the test, but could have an impact on the results or make a difference in a biased manner were classified as control variables. Some control variables used were company size (SIZE), leverage (DER), proportion of independent directors (BIND), board diversity (BDIV), and board size (BSIZE). Company size (SIZE) was measured by calculating the natural logarithm of total assets. Leverage was measured using Debt to Equity Ratio (DER). The proportion of independent directors (BIND) was calculated by dividing the number of members of the board of directors from outside the company by the total number of directors multiplied by 100%. Board diversity (BDIV) was measured using a dummy variable that was equal to 1 if a female director was present on the board and a dummy variable was equal to 0 otherwise. The size of the board of directors (BSIZE) was measured by calculating the number of directors in the board of directors of company i at period t.

Descriptive Analysis
Based on Table 2, audit quality (AC) variable consisting of industry specialist auditors and non-specialist auditors had an average value of 0.31 (31%), which indicated that of the 1,213 firm-years of the sample studied, there were 381 sample companies that selected industry specialist auditors and 832 sample companies that selected nonspecialist auditors as their company's auditors.
The PCON variable in the study had an average value of 3.48, meaning that on average, each company had at least a score of 3.48 for politically connected commissioners and directors. The average DACC value of 0.14 indicated that the average level of earnings management (discretionary accruals) by companies was 14%. The company size variable (SIZE) in the study had an average value of 28.61. The DER variable in the study had an average value of 0.59, meaning that on average, 59% of the total assets of the companies were financed by debt.
The proportion of independent directors (BIND) in the study had an average value of 0.20, where averagely, 20% of the companies had an independent board of directors in its ranks. The representation of female directors in the company's board of directors (BDIV) in the study had an average value of 0.43, meaning that on average, 43% of the sample companies had female directors on the board. The variable of size of the board of directors (BSIZE) in the study had an average value of 4.74, meaning that the average size of the board of directors in the companies consisted of 4-5 people.  (2019) Annotation: The signs ***, ** and * each represent statistical significance levels of 1%, 5% and 10%.

Research Result
The results of the regression testing of the three hypothesis models using the control variables are summarized in Table 3.
Equation Model 1 revealed that the coefficient of variable of company with politically connected boards of commissioners and directors (PCON) was positive at 0.018 and significant. Hypothesis 1, which states that corporate political connections have a significant negative effect on audit quality, had not been proven. This result is not in line with studies conducted by Cheng et al. (2015) and Habib et al. (2017), but supports studies by Guedhami et al. (2014) and (Narayanaswamy 2013).
The results of this study proved that companies with politically connected board of commissioners or board of directors tended to appoint industry specialist auditors as good-quality auditors that the company needed.
In Equation Model 2 in Table 3, earnings management variable (DACC) showed no significant effect. Meanwhile, in Equa-tion Model 3, the interacting variable between the variables of political connections and earnings management (PCON x DACC) indicated no significant effect. These results indicated that earnings management did not moderate the effect of political connections (PCON) on audit quality (AC). Hypothesis 2 was based on hypothesis 1 (there is a negative relationship between political connections and industry specialist). Because Hypothesis 1 in this study was not proven, the argument of Hypothesis 2 became invalid.
SIZE showed a positive regression result <0.001 and was significant at a significance level (p-value) of less than α = 1%. The results of this study were consistent with research conducted by Cheng et al. (2015) who found that large companies tended to select reputable auditors. This was also supported by Liu et al. (2017), who revealed that company size had a significant positive effect, indicating that large companies were more likely to select high-quality auditors.
DER, BIND, and BDIV each showed a negative and significant regression coefficient at a significance level (p-value) of less than α = 10%. These results indicated that the level of leverage, the proportion of independent boards of directors, and the representation of female directors in the board of directors might discourage companies from decision to select a good-quality auditor. BSIZE showed a positive and significant regression coefficient at a significance level (p-value) less than α = 5%. The result of this study, which stated that BSIZE affected auditor selection, supported the research of Alfraih (2017) and Ianniello et al. (2015). One unit increase in the size of the board of directors lead to an increase in the possibility of selecting good-quality public accounting firm. Alfraih (2017) and Ianniello et al. (2015) argued that a larger board made it possible to select high-quality external auditors to mitigate organizational problems and improve the perception of the quality of published financial statements.

Effects of Political Connections on Audit Quality
The Equation Model 1 in Table 3 revealed that the variable coefficient for companies with politically connected boards of commissioners and directors (PCON) had a significant positive effect. The results of the study did not support Hypothesis 1 nor did the research conducted by Cheng et al. (2015) and Habib et al. (2017), which stated that corporate political connections negatively affected auditor selection. However, the result of this study was in line with Guedhami et al. (2014) and Narayanaswamy (2013). The result of this study that did not support Hypothesis 1 did not support agency theory that implies that companies with political connections, which are closely related to the emergence of agency conflicts, tend to minimize information disclosure (Rodríguez et al. 2007) and tend to report poor quality financial statements (Al-Dhamari and Ku Ismail 2015; Chung et al. 2005).
The result of the study that discovered a significant positive effect can be explained by several reasons. The first reason is that companies with a board of commissioners or directors who have political connections in Indonesia tend to appoint industry specialist auditors because they are considered to be of good quality as a mechanism to reduce agency problems and suspicion of information asymmetry. Companies with political connections tend to appoint industry specialist auditors as a company commitment to show the public that people involved in these companies also improve accounting transparency and convince investors that companies refrain from exploiting their connections to divert company resources in the interest of personal. Auditor of a company is appointed by the board of directors. The board of commissioners must supervise and have the right to give approval to the Annual Operating Plan and Budget submitted by the board of directors.
The second reason is the decision to select qualified auditors can enable politically connected companies to eliminate criticism about the quality of their financial reporting by creating better governance and control. The argument of the result of this study is in accordance with Guedhami et al. (2014) and Narayanaswamy (2013). This argument is also consistent with agency theory in which the agent engages with highquality auditors to limit the possibility of rent seeking. The decision to select an independent external auditor as a monitoring role on behalf of the owner/shareholder becomes an important component of corporate governance because it can help improve the credibility of accounting information provided by management (Abdel-Khalik 2002;Lin and Liu 2009). Balsam et al. (2003) argued that industry specialist auditors offered a higher level of assurance than nonspecialist auditors because of the auditor's specialized knowledge of the industry and accounting in the industry. Therefore, the appointment of auditors with industry specialization will help reduce earnings management. Auditors who specialize in the client industry are expected to provide higherquality audit than non-specialist auditors do (Abbott and Parker 2000).

Profit Management as a Moderating Role of The Effect of Corporate Political Connections on Audit Quality
In Equation Model 3 in Table 3, the interacting variable between variables of political connections and earnings management (PCONXDACC) had no significant effect. It meant that the earnings management variable (DACC) did not moderate the effect of the political connections (PCON) on audit quality (AC). Based on these results, Hypothesis 2 was not successfully supported. Hypothesis 2 was based on Hypothesis 1 (there is a negative relationship between political connections and industry specialist) and because Hypothesis 1 in this study was not proven, the argument of Hypothesis 2 became invalid. The result of Hypothesis 2 did not support the research conducted by Chaney et al. (2011), Cheng et al. (2015, and Hashmi et al. (2018). It meant that companies with political connections in Indonesia that selected industry specialist auditors did not have the incentive to manage earnings and report low quality earnings.

CONCLUSION
In agency theory, as a manifestation of management's accountability to principals, agents are obliged to run the company and produce financial reports. The financial statements describe the condition of the company's financial performance that will be used by principals as a basis for decision making. Management engages with highquality auditors to limit the possibility of rent seeking and supports the quality of good financial statements.
Empirical results revealed that politically connected boards of commissioners and directors in a company had a significant positive effect on audit quality. The results of this study have proven that companies with a board of commissioners or board of directors who have political connections tended to select industry specialist auditors who can support the company's commitment in implementing GCG practices. It meant that corporate political connections were used positively in the auditor selection decision (audit quality). Politically connected companies in Indonesia have proven to be committed to developing businesses in accordance with good corporate governance (GCG). This was evidenced by the participation of several politically connected companies in the Annual Report Award (ARA) and Good Corporate Governance Award (GCGA). It proved that these companies were trying to improve the quality of information transparency in the annual report.
The next result of this study was that earnings management did not moderate the effect of corporate political connections on audit quality because the politically connected companies in Indonesia were committed to implement good corporate governance practices. The implementation of good corporate governance can provide higher quality financial reporting. Thus, good corporate governance can reduce the risk perceived by the public and auditors about politically connected companies regarding business failures and possible reporting errors (Wahab et al. 2011). In addition, good corporate governance can show demand for more comprehensive audits from auditors because companies want the public to recognize that they have reliable financial reports and are not subject to governmen assistance or bailouts in the future (Wahab et al. 2011).
The results of this study provide several contributions. The first contribution is to provide empirical evidence that fot the 2015-2017 period, politically connected companies in Indonesia did not make use of their political connections for bad purpose, for example, preserving and covering company's political interests. This study precisely illustrated that politically connected companies made use of their political connections positively by selecting industry specialist auditors who can support the company's commitment to GCG practices. This was implemented by the company for the sake of corporate accountability and business continuity to realize shareholder value in the long term while still taking into account the interests of other stakeholders based on ethical values and laws and regulations. The results of this study can also provide an overview for investors and other institutions (private, public, or state-owned enterprises) regarding the reasons why politically connected companies appoint industry specialist auditors as external auditors.
Several previous studies in Indonesia that discussed a lot about the effect of political connections were focused on tax aggressiveness and tax avoidance, so the second contribution of this study is to provide other empirical evidence about the use of political connections in the business environment in Indonesia. The third contribution is that this study provides another measurement alternative to the variables of political connection and audit quality that can be used for further research. The limitation in this research is the difficulty in obtaining credible information and data in revealing the business environment of industry specialist auditors and non-specialist auditors in Indonesia that are used in proxy of audit quality. Credible journals that discuss the business environment related to industry specialist auditors and non-specialist auditors in Indonesia are still limited. Thus, this study only used information sourced from foreign literature without adjusting to the context in Indonesia. In addition, the Indonesian Institute of Certified Public Accountants also does not have data or information related to industry specialist auditors, non-specialist auditors, big-10 auditors, and non-big-10 auditors in Indonesia.
Future studies can search for alternative proxies of audit quality that are more suitable and relevant to be applied in Indonesia. Then, further studies can develop the effect of political connections in Indonesia on audit quality and executive compensation. Political connection and its relationship with audit quality and executive compensation are still rarely studied in Indonesia.