Auditor Characteristics and Audit Report Lag: Industry Specialization and Long Tenure as Moderating Variables

2021 Abstract Purpose: This study aims to analyze the effect of audit tenure and auditor workload on Audit Report Lag (ARL) and provide empirical evidence of whether the selection of industry-specialized auditors and audit partners with specific workloads can weaken this relationship. Method: This research was conducted using moderated regression analysis. Selection of the research sample using purposive sampling method, from all public listed companies during 2015-2017, 945 firm years were selected. Finding: This study found that audit tenure moderately significantly negatively affects ARL. Besides, this study can also prove that partners with heavy workloads can lead to longer ARL. However, a long partner-client relationship can weaken the workload and ARL relationship because of the auditor’s more familiarity and information. Based on these results, the characteristics of auditors affect the timeliness of audit reporting. Novelty: This study enriched the literature by finding out how to deal with audit delay effectively, whereas previous research only focused on identifying ARL causes. This research’s implication is expected to provide broader insight to the company regarding several factors that can help companies issue their financial reports on time when the auditors have high workloads and short 13(2),


INTRODUCTION
The impact of delay on audit reporting (ARL) on the timeliness of the release of financial reports and investor reactions to accounting information publications has attracted researchers and practitioners' attention. The timeliness of the issuance of financial reports can affect uncertainty in making economic decisions. ARL is an interesting variable because it is the main factor that ensures the release of the company's financial statements is not delayed (Pizzini et al., 2015). If the release time is late, the information gap in the market increases, and financial information's relevance decreases (Bamber et al., 1993;Whitworth & Lambert, 2014). The relevance of information in financial statements is achieved if disclosed to the public on time. The delay in submitting financial reports can cause its ability to influence user decisions (Abdillah et al., 2019). The timeliness of completing the audit is very important because it determines the timeliness of the release of the financial statements (Johnson, 1998). The fact shows that the market reaction worsens when there is a delay in submitting its financial statements (Bartov et al., 2011;Li & Ramesh, 2009). The possibility of this delay is driven by the time to complete the audit (Dunne et al., 2007), indicating that the long audit tenure gives a negative signal.
ARL cases also occur frequently in Indonesia. In 2018 there were more than 24 issuer who had not submitted financial reports. The issuer should submit financial reports no later than three months after the reporting date. Due to this delay, these companies were immediately warning to submit a report (Ayuningtyas, 2019). Since September 30, 2003, Bapepam has tightened its regulations by issuing an attachment to the Decree of the Head of Bapepam Number: Kep-36 / PM / 2003. The regulation states that the submission of audited financial statements must be made at least three months or ninety days after the balance sheet date. The time limit is given to maintain the relevance and reliability of the information required by business actors in the capital market. So it is important to analyze the factors that can shorten the audit report lag in Indonesia.
Delay in audit reports (ARL) can be caused by the company's characteristics (auditee) or from the side of the characteristics of the auditors. Lee et al. (2008) and Hay et al. (2006) found that client complexity is positively related to audit costs, which is positively related to ARL. ARL is indirectly related to corporate governance because its condition affects the planned audit procedures and auditors' control risk assessment (Cohen & Hanno, 2000). Ilaboya and Christian (2014) found that corporate governance affects ARL. Much of the previous literature has shown that individual auditors' characteristics determine the timeliness of the release of audited financial statements (Bamber et al., 1993;Dao & Pham, 2014;Robert Knechel & Sharma, 2012;Whitworth & Lambert, 2014). This study enriches the literature by analyzing whether audit tenure can reduce delays (ARL) due to high audit workloads. Besides, this study also analyzes the interaction between auditor specialization and short audit tenure on ARL. Muhammad (2020) said that a longer auditor tenure could significantly reduce the delay in audit reports. Asni et al. (2017) provide empirical evidence that auditors' educational background and tenure significantly affect audit report lag. Enactment of regulations regarding audit engagement restrictions encourages delays in audit reports (González et al., 2015). Wiyantoro and Usman (2018) also suggest that audit tenure is negatively related to ARL. Abdillah et al. (2019) research show that audit tenure does not harm audit reports delay. Professionalism must be exercised by all auditors of public accounting firms in completing audit work in order to achieve the target completion time and not reduce the usefulness of financial reports as a basis for decision making. Following the compliance theory, individuals or organizations such as auditors must complete the audit process promptly. Manurung et al. (2018) also found that audit tenure does not affect ARL. The difference in previous research results may be due to not exploring long and short tenure conditions. Therefore, this study attempts to fill the research gap by analyzing short-tenured audit conditions. The emergence of mandatory rotation rules for audit firms has prompted rejection reactions because it is feared that it will affect the cost of changing auditors. Changes in auditors may affect audit quality because they are unfamiliar with their client and the client's industry during the initial years of the audit engagement (Lim & Tan, 2010). Another negative impact that is concerned about due to mandatory rotation is the longer the audit delay. Then, The short audit work period can lead to retardment in the information provided to company stakeholders due to the auditor's lack of understanding of their operations (Habib & Bhuiyan, 2011).
It is important to explore how to abbreviate audit report delays in short tenure conditions. Industry specialist auditors are believed to be more capable of detecting errors and completing audits faster than non-specialist auditors because they have special experience and expertise in certain industries. In the end, this capability will drive efficiency improvements in audit assignments (Owhoso et al., 2002). Previous research has shown that public accounting firms with industry specialities can improve audit quality (Balsam et al., 2003;Eriandani et a., 2020;Kwon et al., 2007;Mayhew & Wilkins, 2003). Based on the description above, it is feared that a short audit tenure will cause the ARL to get longer. On the other hand, specialized auditors are believed to shorten audit report delays, so this study suspects that the interaction between shorttenured audit and auditor specialization will reduce ARL.
An overburdened engagement partner is responsible for multiple audits and may not provide sufficient time to direct the audit effort and interpret audit evidence. The relationship between workload and audit report lag is still very limited. Research by Hussin et al. (2018) found that busy auditors are worried about giving inappropriate audit judgments and decisions because they cannot carry out adequate supervision, are in a hurry to gather reliable evidence, and are under pressure due to very limited time. As a result, they are often unable to complete audit engagements promptly. In contrast, Goodwin dan Wu (2016) argues that partners who have many clients are considered more credible, and an increased partner reputation allows them to attract more clients. Other studies mostly analyze the relationship between auditor workload and audit quality. Auditor workload pressure is negatively related to accrual quality and the possibility of a modified audit opinion, as well as the prospect of audit report delay (Chen et al., 2020). López dan Peters (2012) also state that high workload pressure encourages poor audit quality and abusive behavior in individual auditors.
In contrast to previous studies, this paper's originality is to add factors that are thought to reduce the delay in audit reporting. When the workload is high, the auditor must have a long tenure to complete the audit work on time. The tenure measurement is analyzed by the audit firm's tenure, not the auditor because the rotation rule applies in Indonesia. This research provides three contributions. First, enrich the literature by analyzing whether the interaction between auditor specialization and short tenure can shorten the audit report lag. In comparison, previous research has primarily focused on identifying ARL determinants without going any further to reduce audit delays effectively. This study confirms that specialized auditors are a factor that can reduce the negative impact of short audit tenure on ARL. Second, our findings can provide some consideration for regulators in resolving disagreements over mandatory rotation. Third, add to the literature on whether long tenure can shorten ARL due to high auditor burdens.

Audit Tenure, workload, spesialisalized auditor, and Audit Report Lag
The effectiveness and efficiency of the audit performance reflect the quality of the audit. Several studies use audit report lag as a measure to analyze the effectiveness of the audit (Bamber et al., 1993;Newton & Ashton, 1989;Knechel & Sharma, 2012;Tanyi et al., 2010). The audit client will bear additional costs due to the emergence of auditor change obligation rules. Furthermore, audit rotation can lead to delayed delivery of information to the market. In other words, as the tenure of the auditor increases, the ARL will decrease, indicating that the long tenure helps the auditor complete the assignment efficiently (Lee et al., 2009). In the audit assignment process, each audit firm needs time to build a deep understanding of the characteristics and business processes of the company being audited; such as studying the accounting system applied, the internal control system, for example, through standard operational procedures (SOP), and of course all the company's operational activities will be audited. Caramanis & Lennox (2008) provide empirical evidence that audit delays generally occur in short tenure conditions because auditors need more time in the initial years of the engagement. The longer the engagement period with a company, the client's business knowledge and understanding will also increase (Lee et al., 2009). The ongoing relationship between the audit firm and its clients can shorten the ARL period (Dao & Pham, 2014). So that the first hypothesis proposed by this study are: H 1 : The longer the Audit Tenure, the shorter the Audit Report Lag As explained in the previous section, the longer the engagement period, the faster the completion of the audit report. In other words, if the audit tenure is short, it will cause the ARL to be longer. Audit efficiency can be seen from the length of the ARL, measured by the number of days between the audit reporting date and the end of the company's fiscal period (Habib et al., 2019). The longer the ARL, the less relevant the company's financial statements will be, and investors will not receive timely information to increase uncertainty (Bamber et al., 1993). Specialized auditors are believed to reduce ARL so that audits can run effectively (Abernathy et al., 2017). An auditor can be an auditor with industry specialization if the auditor has often audited companies in a certain industry so that he has experience in that field. Auditors with industry specialties are judged to be faster in detecting risks, errors, or abnormalities in the client's business in a certain industry, which can shorten the lifetime of ARL (Habib & Bhuiyan, 2011). Dao & Pham (2014) stated that auditors with industry specialties could alleviate the adverse effects of short audit tenure on the ARL. Short audit tenure resulted in longer ARL, while specialized industry auditors were able to reduce delays. Therefore, this study suspects that the audit will still be carried out effectively, even though the audit tenure is short. In other words, the interaction of industry specialization with short tenure can reduce ARL. So that the second hypothesis proposed by this study are:

H 2 : Auditor Industry specialization mitigates the effect of short audit tenure on Audit Report Lag
This study assumes that overburdened auditors responsible for multiple audits cannot provide sufficient time to direct audit efforts and interpret audit evidence. As a result, they may exercise inadequate oversight and make rush decisions to gather reliable evidence, leading to deterioration in audit quality. Suzuki dan Takada (2016) document that a decrease in audit quality due to high workload can be mitigated by a longer tenure and composition of the audit team. Lai et al. (2018) supports these results, it is also found that the more committed partner assignment resulted in deteriorating earnings quality. Previous literature explains that auditor workload can be viewed from two different sides (Goodwin & Wu, 2016;Gul et al., 2017;Sundgren & Svanström, 2014). On the bright side, auditing multiple clients show that partners are considered as more professional and independent. On the other hand, busy auditors cannot exercise adequate oversight and tend to make decisions hastily to gather reliable evidence due to time pressures, leading to disruption of audit judgments and decision making. Thus, audit partners with heavy workloads tend to spend more time completing their audit assignments. Wan Hussin et al. (2018) show that heavy partner workloads extends audit reporting delay (ARL). So that the third hypothesis proposed by this study is: H 3 : The higher the partner's workload, the longer the Audit Report Lag Auditor characteristics are one of the main variables affecting audit efficiency, which is reflected in the audit report's timeliness. Hussin et al. (2018) argue that long audit tenure is expected to reduce audit effort from engagement partners because information gaps between auditors and partners are fewer. Based on the regulation regarding the obligation to perform rotation, it is estimated that the partner's workload must have a devastating impact on more severe audit reporting delays. This is in line with the results of research by Gul et al. (2017), who found the same thing. However, the researchers observed that partners engaged with more clients in a reporting period reduced the impact of time lag on audit reporting (ARL) when they have had longstanding engagements with their clients. The specific knowledge about clients that accumulate over a long period of audit tenure can reduce busy partners' adverse effects on ARL. So that the fourth hypothesis proposed by this study is: H 4 : Audit tenure weakens the influence of partner workload on Audit Report Lag

METHODS
This study uses a quantitative approach. The population used is all companies listed on the IDX in 2015-2017, except for the financial and banking sectors. The industry is excluded because it has its characteristics and regulations. The selection of research data using purposive sampling method by determining several criteria. Based on table 1, it can be seen that the total population from 2015 to 2017 was 1,374 companies. The companies that met the criteria for the specified research sample were 945 firm-year. The sampling criteria are shown by table 1.
The dependent variable in this study is ARL which can be measured based on the number of calendar days from the end of the fiscal year to the date of issuance of the auditor's report (Dao & Pham, 2014). Audit tenure, auditor specialization, and AP workload are independent variables. One of the factors that affect the efficiency of the firm's audit is tenure audit.
TEN: The number of years of audit engagements between clients and the same partner, tracing back to the research years. Auditors with industry specialization are auditors who have a market share of at least 15% of clients in a certain group. The auditor's industry specialization is measured using a dummy variable. If the Public Accounting Firm controls (KAP) 15% or more, it will be given code one and code 0 if it does not reach 15% (Balsam et al., 2003 (16) (20) (12) (49) The total sample used in the study 316 316 313 945 Source: secondary data processed SPEC = Total assets of KAP clients in the industry The total assets of all companies in the industry Auditor workload is the number of clients that are the responsibility of a partner as measured by a formula (Goodwin & Wu, 2016): LnBussy = Natural log of the number of clients handled by a partner in one year. Referring to previous research (Dao & Pham, 2014;Lee et al., 2009), several factors that influence ARL are included in the research model as control variables. Return on Asset (ROA) is a ratio that measures the return rate of a company's assets. It is calculated by calculating the ratio between net income and total assets. Leverage (LEV) is the amount of debt used to finance company assets. The leverage value is obtained by comparing total debt to total assets. Big Four (Big4) is the top 4 international accounting professional services firms in the world. If Big4 audits the company, it will be coded one and coded 0 if not (Dao & Pham, 2014). Size (SIZE) looks at total assets as a tool to measure the size of a company, where size = natural log of total assets. Going Concern (GC) is an opinion issued by the auditor in the audit report when there is a problem of uncertainty regarding its ability to maintain its business continuity (ISA 706). If the company receives a going concern opinion, it will be given code 1 and 0 otherwise. Financial Restatement (RESTATE) is an activity of reporting back on a company's financial statements. If the company reports back in the year included in the research period, it will be given code 1 and 0 otherwise.
The data analysis technique used multiple linear regression. Before using least squares regression (OLS), a classic assumption test was carried out as a prerequisite. Model 1 is used to answer hypotheses 1, 3, and 4. While model 2 is used to answer hypothesis 2. Testing the second hypothesis is analyzed with a separate model because it analyses the impact of specialization on short audit tenure relationships and audit report lag (STEN * SPEC). Significance levels were seen at one percent, five percent, and ten percent. ARL = α 0 + α 1 TEN + α 2 SPEC + α 3 LnBUSSY + α 4 TEN*SPEC + α 5 TEN*LnBUSSY + α 6 ROA + α 7 LEVERAGE + α 8 BIG4 + α 9 SIZE + α 10 GC + α 11  Notes : ARL : The number of calendar days from the end of the fiscal year to the release date of the auditor's report TEN : The number of years the client has been audited by the same partner, with tracing backwards from the year of study STEN : 1, If the tenure is less than or equal to 3 years, and 0 otherwise. SPEC : Total assets of audit firm clients in the industry / total assets of all companies in the industry. 1, if the audit firm controls 15% or more, and 0 otherwise. LnBUSSY : Natural logarithm of the number of clients audited by partners in a given year ROA : Net earnings / total asset LEVERAGE : Total debt / total asset BIG4 : 1, if the company is audited by Big4, and 0 otherwise SIZE : Firm size, measured by the natural logarithm of total assets GC : 1, if the company accepts a going concern opinion, and 0 otherwise RESTATE : 1, if the company restates its financial statements, and 0 otherwise

RESULTS AND DISCUSSION
Descriptive statistics show the distribution of sample company data, namely the average, minimum value, maximum value, and standard deviation. Besides, this statistical frequency is used to analyze the frequency for dummy variables. Table 2 shows that the number of samples used was 945 samples. The results of the minimum and maximum values show that this study has a wide range of values. The standard deviation value of each variable is quite different from the mean value, and this shows that the values of these variables are well distributed. Table 3 shows that out of a total sample of 945, only 24 companies were audited by auditors with industry specialities. Table 3 shows the sample distribution for auditor specialization, companies that restate, big-four accounting firms, and company going concern conditions. From a total sample of 945, the companies audited by auditors with industry specialities were only 24 companies. There are 744 companies with a value of 0 and 201 companies with a value of 1, which means that of the total sample companies there is only 21.3% of companies restatement their financial statements in the research year. Column Big4 shows the value of 0; there are 637 companies or as much as 64.7% of the total sample of companies, while Big 4 with a value of 1 is 308 or equivalent to 32.6%, which means that there are still not many companies audited by Big4 audit firm. The GC column shows that from the total sample of 945 companies, 2.8% obtained a going concern opinion, which means that almost all sample companies are spared going concern opinion, meaning that there is no indication of the company's uncertainty's ability to continue the business. Table 4 shows the regression results of hypothesis testing, the calculated F value of model 1 of 8,577 with a significance value (Sig.) less than 0.01 (1%). The F value of model 2 is 10,548 with a significance value (Sig.) less than 0.01 (1%). The adjusted R square for model 1 and model 2 is 8.1 percent and 9.4 percent. Based on the results of the t-test in table 4 model 1, the significance of the t value of the TEN variable is 0.08, with a beta coefficient of -0.285. Thus hypothesis 1 is accepted because TEN has a moderately significant effect on ARL, with a significance level of 10 percent. Auditor tenure's negative effect on ARL means that audit tenure has an inversely proportional relationship with ARL, so that hypothesis 1 is accepted. This result is in line with several previous studies stating that ARL becomes shorter when the audit firm tenure (audit tenure) is longer (Dao & Pham, 2014;Muhammad, 2020). These results imply that the audit function at the start of the engagement may be less efficient than subsequent audit work (Aljaaidi et al., 2015). he longer the audit firm engagement period with a company, knowledge and understanding of the client's business will also increase (Lee et al., 2009). In other words, this ongoing relationship between the auditor and his client can shorten the ARL period. In the audit assignment process, every public accounting firm needs time to build a deep understanding of the characteristics and business processes of the company being audited, such as studying the accounting system applied, the internal control system, for example through SOPs, and of course all company operational activities related to the field. The audit to be audited. Thus it can be concluded that the longer the audit tenure, the shorter the ARL will be. This study's results are following the timeliness principle because it has been proven that audit tenure harms ARL. The longer the audit firmclient relationship, the shorter the ARL will be. The longer the engagement period, the audit firm's knowledge and understanding of the client's business will also increase to enable auditors to complete their audits more quickly and report the company's audited financial statement on time.
To answer hypothesis 2, see the results in table 4 model 2, the STEN * SPEC variable shows a t-test significance value of 0.266. Thus, hypothesis 2 is not accepted because STEN * SPEC has no significant effect on ARL. This finding is inconsistent with the research of Dao dan Pham (2014) which states that specialized industry auditors weaken the impact of short-tenured audits on audit report delays. However, this study is consistent with Sawitri dan Budiartha (2018), which states that specialized industry auditors cannot moderate the effect of audit tenure and ARL because there are regulations that require each auditor to rotate. Mandatory rotation for auditors ultimately prevents the auditors from having expertise in a particular industry. Three years is not enough for auditors to obtain sufficient knowledge about the client's business industry, so the potential for delay in the audit report may occur, even though the company has used auditors' services with industry specialities. Based on the regression results, the SPEC variable in model 2 shows a significant value <0.01, while the STEN * SPEC variable is not significant. Thus, it can be interpreted that the industry specialist auditor is an independent variable, not a moderating variable. Furthermore, the specialized industry auditors are not sufficient to minimize the negative impact of short audit tenure on ARL.
The significance value of the t-test for the LnBussy variable in model 1 shows that it is 0.004 with a beta value of 6.947, which means that hypothesis 3 is accepted, LnBussy has a positive effect on ARL. The audit workload variable is proven to have a significant effect on ARL. This study's results are consistent with Hussin et al. (2018), where auditors with multiple assignments will extend their audit reporting delays (ARL). High auditors' busyness causes them to be unable to carry out adequate supervision and tend to make decisions hastily in finding reliable evidence due to time constraints; thus, audit assessments and audit decision making are disrupted. The audit work completion time will be longer if the auditor's workload is high. Besides, Goodwin and Wu (2016) see this condition from two different sides: the advantages and disadvantages of high partner workloads. Advantages auditors with their high workload are deemed to have credibility and independence are more trusted by many clients. On the other hand, auditors' high workload is also a weakness; if there are too many clients, the workload's impact is too large. Furthermore, auditors lack time to perform the audit engagement and interpret audit evidence, which results in a longer ARL. The study results follow the principle of timeliness because partners with large workloads cannot audit carefully because of their limited time. Partners cannot be maximal in conducting audit assignments because of the large number of clients that must be handled, and they do not have enough time to explore their client's business. In the end, it causes the auditor's delay in reporting the company's audited financial statements to the public.
Finally, the t-test significance value of the LnBussy * TEN variable in model 1 is 0.005, with a beta value of -2.066. This means that hypothesis 4 is accepted; TEN can weaken the influence of LnBussy on ARL. The interaction variable between audit tenure and workload is proven to have a significant negative effect at the 1% level, which means that long audit tenure can mitigate the negative impact of partner workloads on ARL. This finding supports the research of Hussin et al. (2018), which states that long audit tenure is expected to reduce information asymmetry so that audit efforts from partners are lighter. This study indicates that if the auditor understands the client's operational flow, then the audit reporting can be on time even though there are quite a lot of clients handled by the auditor. The knowledge and skills developed through repeated audits help reduce the learning curve, which can ultimately ease partners' heavy workload. In other words, partners with more clients handled can still reach the target reporting time when they have had longstanding engagements with their clients.

CONCLUSION
This study aims to identify several factors that can reduce delay in audit reports. The results showed that long audit tenure was able to reduce the delay time in audit reports. Conversely, the shorter the audit tenure, the longer the ARL, and this effect cannot be mitigated by industryspecialized auditors. Another finding is that partner workload affects ARL. Partners with heavy workloads make ARL longer life because they have to handle so many clients at one time. However, the high partner workload's negative impact can be minimized when the partner-client has a long-standing engagement. The partner has special knowledge accumulated during the long audit tenure period so that the ARL period can be shortened.
This study has several limitations: first, the client workload is measured by considering only clients listed on the IDX, so that this study may underestimate the actual workload of partners. Second, this study measures industry auditors' speciality by looking at the total assets of clients in a particular industry compared to all issuers' total assets in that industry. This calculation does not adequately describe the specialization aspects of the auditor industry. Third, this study only analyzes the characteristics of auditors. Future research can include aspects of company complexity and audit quality of client financial statements as factors that influence ARL.