The Effect of Audit Fee, Auditor Rotation, Auditor Firm Reputation, and Auditor Specialization on Audit Quality

This research analyzes the effect of audit fees, audit switching, audit firm reputation, and auditor specialization on audit quality. Data in the current research were secondary data obtained from the financial statements of consumer good industry companies listed on the Indonesia Stock Exchange in 2017-2019. This study used the purposive sampling method to select samples from the population. Based on the method, there were 26 companies obtained with 78 samples. Data were analyzed using multi-linear regression analysis. Proxy measurements for audit quality are by discretionary accruals of the Kaznik model (1999). The result indicates that audit switching has an impact on audit quality. While audit fee, KAP reputation, and auditor specialization do not impact audit quality. This is an open access article under the CC BY-SA license.


INTRODUCTION
Companies enlisted in the Indonesia Stock Exchange (BEI) must issue their audited financial statement. Public accountant service is crucial to measure the soundness of a company's financial statement. The financial statement describes the company's performance, which eventually is helpful beneficial for related parties, both internal and external. A public accountant must maintain the trust and reputation he/she gains from the public.
An auditor holds high integrity because financial audits require an independent and neutral auditor as a part of the responsibility in maintaining public trust. There is an urge to maintain trust from clients and financial statement users; otherwise, auditors will not have the capacity to serve clients and the public in general. According to De Angelo (1981), audit quality opens all possibilities during auditing. It might generate findings that are not always suitable with the client's expectation regarding violation of the accounting system; nevertheless, the report must be made relatively and reported accordingly. Therefore, auditor must subject to the prevailing accounting standard and code of ethics. The low audit quality has drawn much attention as it involves public accountant that should remain independent.
The independence of an auditor determines the audit quality provided. Nevertheless, many cases are still occurring related to the low quality of audit, which eventually disservice the financial statement's users severely.
The latest phenomenon related to audit quality occurs to PT Tiga Pilar Sejahtera Food The development of audit quality is affected by both intenal and external factors. The internal factors included in this study are audit fee, auditor rotation, KAP reputation, and auditor specialization. Firstly, the audit fee is assumed to increase the auditor's performance in completing the work and meeting the client's expectations. previous study Permatasari & Christina (2018) stated that audit fee has a significant positive effect on audit quality. High audit fees can increase the quality and competence of the audit. study is in line with Kurniasih dan Abdul (2014) stated that the audit fee significantly affected the audit quality. Secondly, auditor rotation is necessary to avoid attachment on the client that might affect the auditor's independence, which eventually might cause fraudulence. A study by Permatasari dan Christina (2018) proposed that auditor rotation significantly positively affected audit quality.
Thus, a mandatory audit rotation can affect the audit quality and significantly impact the independent audit result.
Thirdly, KAP's reputation is closely related to good company image, and there is an urge to maintain its reputation before the public consistently. According to Bigus (2015), audit service companies with good reputations have higher motivation to reduce errors and faulty than those with less good reputations. A study by Prasetya and Rosmita (2016) proposed that KAP's reputation positively affected audit quality; the higher reputation a KAP has, the better quality its audit for its clients. This study is supported by Ramadhan (2018) that suggested the KAP reputation significantly affected audit quality.Fourthly, according to Ishak et al., (2015), auditor specialization refers to a particular auditor with better knowledge and understanding of the company's internal control, business risk, and audit risk on the particular industry. A study by Panjaitan dan Anis, (2014) proposed that auditor specialization significantly affected audit quality.
Several studies have been conducted to investigate empirical evidence to affect audit quality. One of them is by Permatasari dan Christina (2018) on audit fee, auditor rotation, and KAP reputation on audit quality. The result of the study proves that audit fees, auditor rotation, and KAP reputation have a significantly positive effect on audit quality. Additionally, a study by Nurintiati dan Agus (2017) on KAP tenure, size, auditor specialization, and audit fee showed that KAP tenure and audit fee had a significant effect on audit quality; on the other hand, KAP size and auditor specialization did not significantly affect audit quality. Another study by Ishak et al., (2015) on audit rotation, workload, and auditor specialization proved that audit rotation negatively impacted audit quality, while workload and auditor specialization positively affected audit quality.
According to the mentioned findings, there is an urge to investigate more factors that may affect audit quality. The study's problem is the effects of audit fee, auditor rotation, KAP's reputation, and auditor specialization on audit quality; these are to find out any impact of independent variables on the independent ones. The result of the study is expected to add insight, knowledge, and understanding within the accounting field, especially concerning audit quality, which eventually serves as a useful reference for future studies.

Population and Samples
The population of this study are manufacturing companies in the consumption goods

Audit Fee
Audit fee refers to the payment received by auditors for their professional services based on such factors as the complexity of the services, the level of expertise, and many other factors. In this study, the calculation of audit fees applied the natural logarithm (Ln) from professional fees received by an auditor (Permatasari dan Christina, 2018).

Auditor Rotation
Auditor rotation is the change of public accountant/ public accountant office to promote independence and avoid attachment with the client, resulting in the decrease of audit quality (Permatasari dan Christina, 2018). Auditor rotation is calculated by using a dummy variable. A company that conducts the rotation will receive 1 point, while those who do not will receive 0 points (Anas et al., 2018).

KAP Reputation
Reputation refers to public trust. When a company experiences a trust crisis from the public, it will negatively impact its reputation; consequently, it requires hard effort to rebuild the trust (Hartadi, 2009). KAP reputation is calculated by using a dummy variable. Therefore, when the KAP audits a Big Four company, it will receive 1 point, and when it is not included in the Big Four, the KAP will receive 0 points (Darya dan Puspitasari, 2017).

Auditor Specialization
Auditor specialization refers to an auditor's capability that includes adequate knowledge, understanding, and experience in auditing financial statements. Auditor specialization is proxied by using the auditor industry specialization (AIS).

AIS = The number of KAP clients in a particular industry The number of isuers within the industry x 100%
Specialization is calculated by using a dummy. Therefore, when an auditor has audited 15% of the total companies by the least, he/ she will receive 1 point; when it is less than 15%, he/ she will receive 0 points (Ishak et al.,2015).

Data Analysis Method
The data analysis technique applied in this study is a quantitative method using multilinear regression, an analytical tool to measure the effect of more than one independent variable on the dependent variable.

Descriptive Statistics Analysis
Descriptive statistics is a method related to collecting, summarizing, and presenting to a more informative form. The descriptive statistics method aims to explain and describe the characteristics of a group of data or more. This method's detailed explanation requires the number of samples, minimum value, maximum value, average value, and deviation standard (Ghozali, 2018).

Classical Asumption Test
This study employed the classical assumption test consisting of (1) normality test to investigate the type of distribution between cofounding/ residual variables using the requiring a criteria Asymp.Sig (2-tailed) > 0,05 will result in the normal distributed data.
(2) The multicollinearity test aims at testing the correlation in the regression model among independent variables; there is no multicollinearity upon VIF<10 and tolerance value >0.01.
(3) The heteroscedasticity test assesses the variance inequality of residual in the regression model using lesser test; heteroscedasticity does not occur upon significance (sig)>0.05;and (4) The autocorrelation test aims at testing the correlation between cofounding in period t and t-1 using Durbin Watson method; there is no autocorrelation indication upon dU < dW < 4-dU (Ghozali, 2018).

Hypothesis Testing
The hypothesis testing in this study applied the determination coefficient (R 2 ), F-test, and t-test. The R 2 is to calculate the model in explaining the variation in the dependent variables. A small value of R 2 means that the independent variables have limited ability to explain their variability. The determination coefficient value ranges from 0 to 1. (2) The Ftest investigates the independent variables' ability to explain the dependent variables; in other words, it is for the Goodness of Fit test (Ghozali, 2018). The significance value <0.05 shows the effect on the dependent variable, while the significance value>0.05 shows the absence of effect on the dependent variable. (3) The t-test investigates the impact of independent variables individually in explaining the dependent variables (Ghozali, 2018). The significance value <0.05 shows a significant impact on the dependent variable, while the significance value>0.05 shows the absence of significant impact on dependent variables. Based on table 5, the significance value of all independent variables is more than 0.05.

RESULTS AND ANALYSIS
The data in this study have the same variance in the regression function or homoscedasticity.
Thus, there is no occurrence of heteroscedasticity. The audit quality variable measured by the discretionary proxy with 71 samples (N) has a minimum value of -0.01 and a maximum value of 0.01. The audit quality variable has an average value of 0.0016 and a standard deviation of 0.00319. The standard deviation value is greater than the average value, meaning that the total accruals are heterogeneous.  suspects that the number of audit fees paid by the company for the auditor's services will not affect the final results of the audit of financial statements that have been carried out as the quality of the audit is proven by the auditor independence during the audit and financial statement. The auditor independence factor of an auditor is directly proportional to the quality of the audit. Thus, audit quality is not affected by the fees received. Agency theory can address agency problems between parties who work together but have different goals by presenting an independent auditor. The audit fee from the company to the audit does not affect the performance and ability of the auditor in auditing the financial statements; the audit result remains objective. The results of this study support research conducted by Ramdani (2016), stating that audit fees do not affect audit quality as they cannot predict the audit quality.

Hypothesis Results
Furthermore, audit quality depends on an auditor's independence; thus, audit quality cannot be measured using the fees given. However, this finding is not in line with the results of Permatasari and Christina's (2018) research, which states that audit fees affect audit quality.
H2 indicates that the auditor rotation variable has a significant effect on audit quality.
Thus, if the company rotates its auditors, it will affect the company's audit quality. Researchers suspect that the change of auditors can reduce attachment between clients and auditors. Doing so can maintain the auditor's independence and objectivity in auditing the company's financial statements, which eventually increases the audit quality. Government Regulation (PP) Number 20/2015 concerning "Practice of Public Accountants" article 11 paragraph (1) explains that the audit services provision limit to an entity by an auditor is by maximum of 5 (five) consecutive financial years (www.jdih. kemenkeu.go.id). Therefore, enacting regulations requiring an auditor to undergo rotation can prevent an auditor from having a close relationship with his client.
Agency theory in this study can address agency problems, namely a long engagement period between the auditor and the client company that may create emotional closeness. This condition will impact the auditor's independence, resulting in financial statements that do not reflect the actual situation. Consequently, the quality of the audit of the financial statements will be of low quality. Therefore, the auditor rotation policy issued by the government is preferable to help companies reduce agency costs and have a good impact on the resulting financial statements.The results of this study support research conducted by Permatasari and Christina (2018), which states that auditor rotation has a significant positive effect on audit quality because mandatory auditor turnover can affect audit quality. This finding is not in line with research Ishak et al., (2015) (Ishak et al., 2015), which states that auditor specialization positively affects audit quality.

CONCLUSION
This study was conducted to test and analyze audit fees, auditor rotation, KAP reputation, and auditor specialization on audit quality in consumer goods manufacturing companies listed on the Indonesia Stock Exchange (IDX) during 2017-2019 by using the empiricism method. The sample selection employed a purposive sampling technique, resulting in 26 companies were obtained for 3 (three) years; thus, the data for this study were 78 samples. However, after managing the data, seven extreme data must be removed from the research sample. Consequently, the remaining total sample is 71. This study applies agency theory to overcome the information asymmetry between the parties who enter into a contractual relationship, namely the shareholders and the agent, because these parties have different goals.
The ability of the independent variable in explaining the variation of the dependent variable is by 25.8%, and the external factors excluded in this study explain the remaining 74.2%. Data analysis proves that the audit fee has no significant effect on audit quality; therefore, H1 is rejected. This finding means that the audit fee received by the KAP does not affect the final results of the financial statement as the audit quality depends on the auditor's independence. Meanwhile, auditor rotation has a positive effect on audit quality; therefore, H2 is accepted. By that, the change of auditors can reduce the attachment between the client and the auditor, which eventually will improve the audit quality.
KAP reputation has no significant effect on audit quality, which means that H3 is rejected. This finding indicates that affiliation with the Big Four does not ensure that the KAP produces better audit results than those who do not. A qualified audit depends on the auditor in complying with auditing standards and the applicable code of ethics. Finally, auditor specialization does not affect audit quality; thus, H4 is rejected. This finding indicates there is no difference between the specialist auditors and non-specialists as both have the knowledge, experience, and can detect any fraud and irregularities in the financial statements, and can produce qualified audits. Further researchers suggest expanding the research object to all remaining sectors listed on the Indonesia Stock Exchange (IDX). Thus, the future study is expected to generalize the results of research and add independent variables such as workload, as a previous study proposed that the greater the workload will increase the auditor's performance so that it affects audit quality (Wismardany, 2018). Another suggestion is to increase auditors' professional skepticism; thus, auditor constantly questions and evaluates audit evidence critically to generate a qualified audit result (Alfiati, 2017).