Does Audit Matter in Earnings Quality of Indonesia Banks?

Abstract

This study investigates and analyzes the difference in Indonesian banks' earnings quality in the pre-audit and post-audit period. This study also investigates the difference in audit quality done by public accounting firms. This study employs time series data taken from the unaudited and audited financial statements of banks listed on the Indonesia Stock Exchange in 2012-2016. Sample selection is made by using a purposive sampling method. The population of this study is 43 banks, and after checking the data for validity and reliability, the final sample size was 26 banks. Audit quality is operationalized with the size of the auditor. Earnings quality is proxied by accruals calculated using the Beaver and Engel (1996) model. The data analysis method used in this study is the paired-sample t-test and chow test. This study shows that there is no difference in earnings quality in the pre-audit and post-audit period. This study also reveals no difference in audit quality between the big four and non-big four auditors. These findings mean that independent auditors do not play a useful role in increasing the reliability of accounting information presented by management to stakeholders. Besides, this study's results do not verify the agency theory regarding auditors' role to minimize opportunistic management behavior in preparing financial statements.

1. Introduction

Earnings are the information component in the financial statements that are most sought by stakeholders. Earnings are also the main factor in determining stock prices because earnings are useful to consider the prospect of the long-term profitability of a firm. Earnings do not always show accurate information. Earnings can mislead stakeholders because there are two components in earnings information: cash flow and accruals. Incorrect valuation or mispricing occurs when investors cannot distinguish the two components of earnings.

The accrual basis provides more factual information related to the condition of the firms but also provides an opportunity for management to overstate the earnings through the modification of the financial statements. Management seeks to generate earnings based on the wishes of the stakeholders because of specific motivations such as the existence of bonuses that will be given as a form of appreciation for management’s performances as measured by the earnings generated. Earnings management is a matter of concern for organizations because it affects the interests of stakeholders. This reduces the quality of information on financial statements of the organizations when the organization performs earnings management behavior. Therefore, firm managers will use various earnings management techniques to achieve earnings targets. Modification of financial statements is done through earnings management, which can cause a decline in earnings quality (Tran et al., 2020).

Financial Accounting Standards (FAS), which are the adoption of International Financial Reporting Standards (IFRS), have brought significant changes to the financial reporting of all sectors of Indonesia, including the banking sector. Statements of Financial Accounting Standards (SFAS) 50 and 55 emphasize the measurement of inadequate credit reserves or better known as loan loss provision. Before the loan loss provision (changed its name in the issuance of SFAS 50 revised 2010, and 55 revised 2006), the term allowance for earning assets losses was known. Loan loss provision is a component in the banking financial statements, which is now in the spotlight. The financial crisis that occurred due to corporate failure to anticipate credit defaults triggered by sudden changes in economic conditions in 2008 caused loan loss provision to be in the spotlight. Some practitioners stated that credit provision was late and too little so that the standard setters were encouraged to amendment previous IFRS in 2013.

SFAS 50 and 55 are principle-based, providing an opportunity for management to conduct earnings management. Jin et al. (2018), and Lee and Choi (2016) reveal that a loan loss provision is a tool used as an extent for earnings management by management in the banking sector. Determining the amount of loan loss provision requires much consideration from management, which tends to be subjective (Nicoletti, 2018).

Auditors are trusted by stakeholders to provide more reliable earnings information. The audit results are expected to increase the credibility of the financial statements so that they can be used as a decision-making tool (Alqirem et al., 2020). Many studies identified the effect of audit quality on earnings quality, but very few identified the role of audit on earnings quality. In the audit process, the auditors carry out an audit adjustment that is deemed necessary for accounts considered material. However, the audit process also experiences limited time and costs so that the auditors only do a sampling of firm transactions. As a result, auditors may pass representative transactions. Besides, the audit process also uses various other subjective considerations such as the calculation of estimated accounts with low accrual reliability such as depreciation and allowance for impairment losses. Even Arens et al. (2014) stated that auditors also can make mistakes even though they are in good faith and integrity, so with the existing audit limitations, there is no guarantee that auditors will play a role in improving earnings quality. Two possibilities can occur namely the financial statements made by the auditor can have a high quality of earnings because of no accounting adjustments, and the financial statements have high earnings quality because the presentation is reasonable and following the applicable accounting standards.

Auditors may not be effective in increasing the credibility of financial statements in practice. The relationship between the auditor and the client is unique in that it can cause a conflict of interest. Auditors may need adjustments to correct client earnings misstatements, but they must consider the consequences. The consequence is that adjustments made can provide a good reputation for auditors, but will also reduce auditor costs (Amat & Elvira, 2010). The study by Amat and Elvira (2010), and Lennox et al. (2016, 2018) showed that audit adjustments affect the earnings quality of a firm. Meanwhile, research conducted by Choudhary et al. (2018) and Hasanah and Suzan (2015) found that audit adjustment does not affect the quality of bank earnings, so there is no difference in earnings quality between before and after the audit. These previous studies show inconsistent results when testing the role of audits in improving earnings quality.

In addition to knowing the role of audits in improving the quality of bank earnings, the effect of audit quality on improving earnings quality needs to be identified as part of the specific role of the audit. Many studies have tried to measure audit quality with various models, such as the size of public accounting firms. Larger public accounting firm sizes can provide higher quality audit services. Larger public accounting firms have a better reputation that needs to be protected, besides many training programs are provided to their auditors and staff, audit methods used are more accurate and standardized, and there are more possibilities in reviewing audit results (Dopuch & Simunic, 1980).

Lee and Choi (2016) found that high-quality audits can improve the reliability of the accruals by reducing accrual measurement errors through the collection and evaluation of more and better audit evidence by using the competence and professional judgment of an auditor. Balsam et al. (2003) stated that improving audit quality will increase investor responses to reported income. Besides, audit quality describes how well audits detect and report material misstatements from financial statements, reduce asymmetry information between principals and agents, and help protect the interests of shareholders (Salehi & Azary, 2009). Ozili (2017) revealed that the presence of Big 4 auditors did not improve the informativeness of loan loss provisions estimates among banks. These studies also show inconsistent results when testing audit quality in improving earnings quality.

This study is unique compared to other studies, in which few studies evaluate the role of audit in financial statements. The majority of previous studies have focused more on differences in audit quality based on their proxies. Based on the explanation above, the study’s motivation is to examine the role of audit on earnings quality in the Indonesian banking industry.

2. Literature Review

2.1. Agency Theory

Individual behavior sometimes shows opportunistic behavior towards their environment, which is behavior that looks for profit for themselves. The opportunist phenomenon in the business world has been explained by Jensen and Meckling (1976), better known as agency theory. The development of a company or business entity that is increasingly large enables conflicts between the principal and agent. Conflicts between shareholders and management can be minimized by the way management must run the firm under the interests of shareholders, including in making decisions by management. Shareholders also monitor managers in running their firms. In order to minimize agency problems, an independent party is needed to mediate in handling the conflict, which in this case, is the auditor of the public accounting firm.

2.2. Hypotheses

Earnings management conducted by banks can reduce the quality of reported earnings, so financial statements audit is needed. The financial statement audit conducted by the auditor aims to assess the reasonableness of the financial statements presented to obtain reasonable assurance about whether the financial statements are free of material misstatement, whether caused by fraud or error. In forming an opinion on whether the financial statements are presented fairly, in all material respects, in conformity with the applicable financial reporting framework, the auditor should take into account all relevant audit evidence, regardless of whether it appears to corroborate or to contradict the assertions in the financial statements. The auditor makes an adjusting entry when he finds a material misstatement in the auditee’s financial statements. An audit adjustment is a journal entry that is used to correct a misstatement detected during the audit. Audit adjustments are critical to ensuring audit quality because if the misstatements are material and remain uncorrected, then a known misstatement remains in the audited financial statements, which is obviously unacceptable by the users of financial statements. The adjusting entries proposed by the auditor must be approved by the auditee because the presentation of financial statements is the client’s responsibility. Audit’s role in improving the quality of financial statements is not known directly because stakeholders only know the final results of the financial statements. Accruals are accounting concepts which are used to recognize expenses and income of the company that has been consumed/earned within the specified financial reporting period to ensure that the company’s expense and income are recorded and recognized based on actual. The researchers believe that when the auditor has carried out audit procedures properly and thoroughly on accrual accounts that have the potential to be manipulated and conducted audit adjustments to material misstatements, the audit will contribute to improving the earnings quality so that the earnings quality in the pre-audit can improve in the post-audit.

Amar (2014) found that the audit process contributes efficiently to improving financial statements, so there is no bias. This result is also in line with research conducted by Grein and Tate (2011), Lennox et al. (2016), and Hasanah and Suzan (2015) who found evidence that audits appear to reduce potential management bias, particularly to reduce risks of an overstatement. They revealed that management tends to overstate earnings driven by a desire to obtain incentives, bonuses, compensation, or other conflicts of interest. Auditing helps in discouraging earnings management carried out by the firm so that the earnings quality after being audited becomes high. The investigation is based on the following hypothesis:

H1: There is a difference in earnings quality in the preaudit and post-audit of the Indonesian banking industry.

This study evaluates the earnings quality in the pre-audit and post-audit in order to assess the effectiveness of the audit. However, this study also examines whether the results of the audit role in earnings quality apply to all auditors, namely big four and non-big four auditors. Researches conducted by Alzoubi (2016, 2018), and Ozili (2017) found that audits conducted by big four auditors are more able to reduce earnings management in firms compared to audits conducted by non-big four auditors. Big four auditors are often identified with high audit quality because big four public accounting firms have more experience and employees than small public accounting firms.

Grein and Tate (2011) identified potential financial statement areas at risk of manipulation based on incentives specific to public housing authorities. Using pre- and post-audit financial data for almost they found that auditors make economically and statistically significant adjustments to financial statements. In addition, they found that audits appear to reduce potential management bias, particularly to reduce risks of an overstatement. According to Hasanah and Suzan (2015), there is no difference in earnings quality before and after the audit, which means that the audit adjustment does not help improve the earnings quality of banking in Indonesia. These findings are not consistent with the studies of Amat and Elvira (2010) and Lennox et al. (2016). Amat and Elvira (2010) stated that the majority of firms tend to overstate than understate firm profits, so the role of the auditor is crucial to prevent earnings management practices. Besides, Lennox et al. (2016) found that audit adjustments made companies’ profits smoother and more persistent.

High-quality audits (measured using auditor size) can increase the accrual component’s reliability by reducing accrual measurement errors through the collection and evaluation of more audit evidence, as well as better auditor competence and professional judgment (Lee & Choi, 2016). Alzoubi (2018) stated that audit quality (audit tenure, auditor size, auditor specialization, and independent auditor) has a significant and negative relationship to earnings management, and high debt has a positive relationship with earnings management. Slaheddine and Fakhfakh (2017) found that, in general, audit quality affects earnings quality in both the United States and France. Kanagaretnam et al. (2010) found a relationship between banks audited by big five public accounting firms. They found that both auditor type and auditor industry specialization moderate benchmarkbeating (loss-avoidance and just-meeting-or-beating prior year’s earnings) behavior in banks. However, they found that once auditor type and auditor industry specialization are included in the same tests, only auditor industry specialization has a significant impact on constraining benchmark-beating behavior. In contrast, Christiani and Nugrahanti (2014) found that public accounting firm size does not affect earnings management. The existence of big-four auditors is not to reduce earnings management but rather to increase the credibility of financial statements by reducing disruptions so that the financial statements presented are reliable. The same thing was stated by Ozili (2017), who found that the presence of Big 4 auditors did not improve the informativeness of loan loss provisions estimates among banks. The investigation is based on the following hypothesis:

H2: There is a difference in the effect of audit quality on the earnings quality of banks audited by big four and non-big four auditors in Indonesia in the pre and post-audit period.

3. Research Method

This research is a quantitative study that uses secondary data in the form of audited and unaudited banking financial statements from 2012-2016, which are listed on the Indonesia Stock Exchange. The data is downloaded through the official website of the Financial Services Authority (www.ojk.go.id) and the Indonesia Stock Exchange (www.idx.co.id). Sample selection is made by using a purposive sampling method. The total population of this study is 43 banks, and after checking the data for validity and reliability, the final sample size was 26 banks.

The role of the audit in this study is seen from the financial statements before the audit and the financial statements after the audit. Meanwhile, audit quality is operationalized with the size of the auditor, namely the big four and non-big four (Nguyen et al., 2020). Earnings quality is proxied by accruals calculated using the Beaver and Engel (1996) model. The Beaver and Engel (1996) model use the allowance for receivables and loan loss provisions as a component of forming accruals in banking firms.

Hypothesis testing is done in two ways. The first hypothesis is tested using paired-sample t-tests. The second hypothesis is tested using a chow test and paired-sample t-test. The regression equation used in the chow test is as follows:

DA = α1 + β1PA + ε1, for all sample big four and non-big four       (1)

DA = α1 + β2PA + ε2, for all sample big four       (2)

DA = α1 + β3PA + ε3, for all sample non- big four       (3)

where DA is discretionary accrual, α1 – α3 are constants, PA is the role of audit measured by pre- and post-audit period (dummy variable), β1 - β3 are estimated regression coefficients, and ε1 – ε3 are error terms.

4. Results and Discussions

4.1. Results

The first hypothesis in this study was tested using a paired-sample t-test. Previously, this research data had been tested for normality to meet classical assumptions. The result of the normality test using the Kolmogorov-Smirnov test shows that the research data is normally distributed. Meanwhile, the paired-sample t-test results, as shown in Table 1, reveal that the significance is more than the 5% significance probability. This result means that there is no significant difference between discretionary accruals in preand post-audit.

Table 1: Paired-Sample t-Test Result for Hypothesis 1

Hypothesis 2 is tested using a chow test. Chow test results show that RSS3 from equation 1 is 0.632, RSS1 from equation 2 is 0.524, and RSS2 from equation 3 is 0.083. Based on Table 2, 3, and 4, the F calculated is 2.53, which is smaller than the F table 3.07. This result means that there is no difference in the effect of audit quality on earnings quality in the pre and post-audit periods.

Table 2: Result of Regression Equation 1 (Big Four and Non-Big Four Auditors)

Table 3: Result of Regression Equation 2 (Big Four Auditors)

Table 4: Result of Regression Equation 3 (Non-Big Four Auditors)

The results of the chow test are strengthened by the results of paired sample t-tests in Table 5 related to the second hypothesis. The paired-sample t-test results for the big four auditors show that there is no difference in discretionary accruals in the pre and post-audit periods. These results also apply to the non-big four auditor sample. Therefore, the second hypothesis is stated rejected.

Table 5: Results of Paired-Sample t-Test for Big Four and Non-Big Four Auditors

4.2. Discussions

4.2.1. Difference in Earnings Quality in the Pre and Post-Audit of the Banking Industry in Indonesia

The first hypothesis of this study which states that there are differences in the quality of bank earnings in the pre and post-audit periods is not proven empirically. This result does not verify agency theory, which states that one effective way to minimize opportunistic management behavior is the existence of an independent party between the agent and principal relations. In the context of a business entity, an independent auditor is a mediator who minimizes the information asymmetric by examining the financial statements prepared by management. However, this study reveals that independent auditors do not play a useful role in increasing the reliability of information presented by management to stakeholders. This conclusion is indicated by the absence of differences in earnings quality in the pre and post-audit periods.

The results of this study are consistent with Hasanah and Suzan (2015), who stated that there is no difference in earnings quality before and after the audit. These results imply that audit adjustment does not play a role in improving earnings quality. Earnings quality measures in Hasanah and Suzan (2015) differ from this study, as they used earnings variability, earnings persistence, and earnings predictability as earnings quality measures. However, the results of this study are not consistent with Amat and Elvira (2010), Lennox et al. (2016), and Tran et al. (2020). Lennox et al. (2016) examined how adjustments to earnings during year-end audits affect measures of earnings quality. There are four key findings. First, audit adjustments cause earnings to become smoother and more persistent. Second, the adjustments result in higher accrual quality. Third, audit adjustments have a larger negative effect on signed accruals than absolute accruals. Fourth, the adjustments do not reduce the discontinuity in the earnings distribution around zero The results of Amat and Elvira (2010) supported the significant role of auditors in the financial market, in particular in the prevention of earnings management practice. Their findings confirmed that companies more often overstate than understate their earnings.

The results of this study can be contradictory to different research subjects, namely the manufacturing industry and the banking industry since the accounting discretion used is different. Accounting discretion in the manufacturing industry is different from accounting discretion in the banking industry. The structure of transactions in the manufacturing industry is more scalable than in the banking industry because the majority of transactions in the manufacturing industry consist of past events. Whereas banks must take into account the risks that may occur in the future associated with the portfolio of productive assets owned. as such, the measurement and recording process is far more complicated than the manufacturing industry and there is subjective judgment in its determination (Henry & Holtzman, 2006).

4.2.2. Difference in the Effect of Audit Quality on Earnings Quality in the Pre and Post-Audit of the Banking Industry in Indonesia

The second hypothesis of this study which states that there are differences in the effect of audit quality on the quality of bank earnings audited by the big four and nonbig four auditors in Indonesia both in the pre-audit and post-audit is not proven statistically. This result shows that the audit conducted by the big four does not reduce the discretionary accrual in the banking financial statements and does not provide meaningful changes to improve the quality of banking earnings. Likewise, there is no difference in discretionary accruals before and after the audit by nonbig four.

The results of this study are not consistent with the studies conducted by Alzoubi (2016, 2018), Ozili (2017), Slaheddine and Fakhfakh (2017), and Tran et al. (2020). These studies stated that audits conducted by the big four are able to reduce earnings management. The results of this study differ from the findings of the study by Tran et al. (2020) who stated that the big four auditors played an important role in minimizing earnings management in the banking sector. However, this study is in line with the study by Christiani and Nugrahanti (2014), who stated that the size of a public accounting firm does not affect earnings management. However, previous studies do not identify earnings management practices in the pre-audit and post-audit period.

Research conducted by Gray et al. (2011) involving several stakeholders stated that after the accounting scandal committed by Enron that was audited by one of the big six public accounting firms, the Sarbanes Oxley Act was issued. The regulation strictly regulates public accounting firms, which resulted in the big six firms becoming the big four. They investigated perceptions and misperceptions regarding the auditor’s report and conducted focus groups with five different stakeholder groups—financial statement preparers (CFOs), users (bankers, analysts, and non-professional investors), and external auditors. Their findings included that financial statement users value the audit, but do not read the entire auditor’s report. It is not clear to users, preparers, and auditors what the auditor’s report is intended to communicate or the level of assurance being provided by the report. Stakeholders offered numerous suggestions to improve the auditor’s report, but they also recognize those suggestions could significantly change the auditor’s risk profile and increase audit fees. Besides, Nelson (2006) also stated that strict rules increase incentives for non-big four to have the same quality as the big four.

This result is supported by research conducted by Nindita and Siregar (2013), who stated that the supervision of big four auditors by regulators is less strict due to market dependence on the big four. This dependence arises because of the assumption that the big four is superior to the non-big four in terms of their ability to provide quality audit services to large firms. Another reason put forward by Wibowo and Ghozali (2017), was that the big four might not always provide high audit quality because the risk of litigation faced by the big four in Indonesia is relatively small.

The results of this study indicate that audit quality measured using auditor size ie, big four and non-big four, is not always be able to minimize discretionary accruals in the banks and does not provide meaningful audit adjustments to improve earnings quality. According to Tandiontong (2016), differences in quality are proven to occur among the size of the accounting firms, but that does not mean that the size of the accounting firm is audit quality or auditor quality. The case involving Enron is enough to show that the proxy is not always valid. Academics generally agree that quality audits must be carried out by competent and independent auditors (DeAngelo, 1981; Tandiontong, 2016). The auditor is responsible for the accounting profession, professional colleagues, public accounting firms, auditee, himself, and broadly to the public so that regardless of the size of the auditor, that is big four or non-big four, when the public accounting firm has the competence and independence in auditing it will produce high-quality earnings.

5. Conclusions

The results of this study provide empirical evidence that there is no difference in earnings quality in the pre-audit and post-audit period in the banking industry in Indonesia. Other findings support this result, showing that audit quality does not affect the earnings quality of banks in Indonesia. The implication of these results is that agency theory is not verified, which stated the agency cost. Agency costs typically arise in the wake of core inefficiencies, dissatisfactions, and disruptions, such as conflicts of interest between shareholders and management. Based on the findings of this study, regulators should formulate regulations in order to improve audit quality through monitoring and supervision, both at big four and non-big four public accounting firms. Besides, investors and potential investors need to be careful in analyzing banks’ financial statements even though big four auditors audit the banks. Even so, the study’s findings must be interpreted and analyzed carefully, considering the focus of this study is the banking industry. Also, this study does not consider the complexity of measuring the allowance for impairment losses that affect earnings management in banks. Meanwhile, the researchers suggest the next researchers use a more varied measure of earnings quality to enhance the findings of this study.