Audit Quality Effects on Earnings Management of Manufacturing Firms in Nigeria: A Comparative Study of Pre- and Post-International Financial Reporting Standard (IFRS) Period

This study examined the effect of audit quality proxied by Auditors’ Independence (ADI), Audit Firm Size (AFS), Auditor Tenure (ADT) and Audit Firm Specialization (ADS) on shareholders’ earnings (measured as Earnings per Share EPS) and stock performance (measured as Market Price of Stock MPS) of listed manufacturing companies in Nigeria, in the pre-and post-International Financial Reporting Standard (IFRS) periods. To achieve this, the study used eleven (11) listed manufacturing companies listed companies that had consistently published their audited annual financial reports from 2009 to 2018. Descriptive statistics, correlation analysis and Ordinary Least Square (OLS) univariate and multiple regression technique were adopted to analyse data obtained, with the ex-post facto research design employed in the methodology. The following results were obtained from the test of hypotheses. Auditors independence, audit firm size and auditors firm specialization have significant and positive impact on EPS and MPS. However, auditors’ tenure was found to significantly affect both EPS and MPS negatively. The IFRS moderated model results revealed that audit quality variables have higher and more positive impact on EPS and MPS in the post-IFRS period, relative to the pre-IFRS, with their effects being statistically significant using the wald restriction test. The findings have direct implication on earnings management and stock performance in the Nigerian Manufacturing Industry.

Ho2: The effect of audit quality on shareholders' earnings do not significantly differ for the pre-and post-IFRS periods of listed Manufacturing firms in Nigeria 2.0 Literature Review 2.1 Theoretical framework Essentially, auditors' theory of inspired confidence, signaling theory, and agency theory justify the key function of auditing as a mechanism for mitigating information asymmetries among related parties.
Developed by the Limperg Institute in Netherlands in 1985, the theory of inspired confidence states that the auditor, as a confidential agent, derives his broad function in society from the need for expert and independent examination as well as the need for an expert and independent judgement supported by the examinations. Thus, auditors are expected to know and realize that the public continues to expect a low rate of audit failures. This requires that the auditors must plan and perform their audit in a manner that will minimize the risk of undetected material misstatements. The accountant is under a duty to conduct his work in a manner that does not betray the confidence which he commands.
The significance of the theory of inspired confidence is that the duties and responsibilities of the auditors are a derivation from the confidence that are bestowed by the public on the success of the audit process and the assurance which the opinion of the accountant conveys. Since this confidence determines the existence of the process, a betrayal of the confidence logically means a termination of the process or function. Carmichael (2004) in discussing the social significance of the audit stated that when the confidence that society has in the effectiveness and quality of the audit process and the audit report is misplaced, the value relevance of that audit is destroyed. Therefore, auditors are expected to maintain reasonable quality assurance especially given that an audit failure is effectively a career-ending event. Audit provides assurance to the owners and management of companies and to investors and stakeholders, and along with financial reporting, corporate governance and regulations, supports confidence in the capital markets.
Signaling theory suggests that companies with good performance use financial information disclosure to send signals to the market. Craven and Marston (1999), show that firms will attempt to accept the same level of disclosure as similar firms operating in the same industry since if a firm does not keep up with the same level of disclosure as others, it may be perceived by stakeholders that it is hiding bad news or negative information. As the types of financial statements produced have become standardized, potential information differentiation that a company can use to send a signal to the market through its financial statements is reduced. Companies are thus provided an incentive to signal, other than through transparency in their notes to the accounts and other voluntary disclosures, through their choice of auditor. Moreover, even voluntary disclosures that may be used as signals achieve enhanced credibility in the presence of a quality auditor.
Thus, a high quality audit sends a signal to the market that the financial statements are more credible than those audited by lower quality auditors. The market perceives audit firm size and specialist auditors to be of a higher quality than others and rewards (punishes) companies with larger improvements (or falls) in stock prices accordingly (Teoh and Wong, 1993;Krishnan & Yang, 1999;Menon and Williams, 1994). It has been shown that the market's perception of the quality of the company's auditor influences that company's stock price. As such, directors and management may want to signal to the shareholders that their interest is being well monitored. Therefore, signaling should, theoretically, affect the demand for audit quality over and beyond the monitoring function alone. The positive Signal of transparency and credibility it sends to the market and the assurance it provides to stakeholders about the quality of earnings performance disclosures suggests a positive association between EPS and audit quality.
Again, signaling theory would suggest that the provision of voluntary IFRSs disclosures would give an indication of firms' decision-making process and financial behaviour (Eccles, Hertz, Keegan & Phillips, 2001). For example, voluntary IFRSs disclosure may signify the intention of firms to distinguish themselves and give positive signals to market participants about their managerial ability and performance (Watson, Shrives & Marston, 2002). The adoption of IFRS gives a positive signal of higher quality accounting and transparency (Tendeloo & Vanstraelen, 2005) and would also lead to lower information asymmetry and cost of capital (Leuz & Verrecchia, 2000). Similarly, the provision of quality accounting disclosures would tend to reduce the opportunities for earnings manipulation and enhance the stock market efficiency (Leuz & Wüstemann, 2003). The higher disclosure requirements and financial reporting quality that stem from IFRSs implies that the adoption of IFRSs gives a positive signal to investors as information asymmetry and agency costs tend to diminish (Tarca, 2004).

Conceptual Review
Since the mandatory adoption of IFRS by a lot of countries in the world, a lot of effort has been put in research to determine the effect it has on key variables such as value relevance and earnings management. This section thus reviews the relevant theoretical and empirical literature in connection with the topic under study.
There are generally two views of earnings management. Firstly, it is seen as a management tool used for fulfilling their responsibility of wealth maximization. Secondly, earnings management is considered as an alteration of economic events in order to mislead the users of financial statements. These two different perspectives have led to various definitions of earnings management in prior literature.
On the aspect of earnings management having a positive effect, earnings management is defined by Beneish (2001) as a way for managers to disclose their private expectations about the firm's future cash flows to investors. This was also supported by McKee (2005) who defined earnings as reasonable and legal management decision making and reporting intended to achieve stable and predictable financial results. Fields, Lys and Vincent (2001) were also in support of this view. On the perspective of earnings management having a negative effect, the following definitions are put forth.
Value relevance, on the other hand, is one of the proxies outside earnings management used in measuring financial reporting quality. According to Vishnani and Shah (2008), value relevance depicts the ability of accounting information in financial statements to explain how the stock market is measured. It is also the ability of information to capture of affect share value (Barth et al., 2001;Hellstrom, 2009 (IAASB, 2011), there has been a number of attempts to conceptualize "audit quality" in the past. However, none has resulted in a definition that has achieved universal recognition and acceptance. Audit quality is, in essence, a complex and multi-faceted concept. The classic definition of audit quality that is cited by most audit researchers is that of DeAngelo (1981) which states that audit quality is the market-assessed joint probability that a given auditor will both (a) discover a breach in the client's accounting system and (b) report the breach. The definition highlights two important aspects of audit quality: (1) the competence of the audit firm that determines how likely it is that a misstatement will be detected and (2) the independence and objectivity of the auditor that determines what the auditor is likely to do about a detected misstatement. This definition has been quite useful to audit quality studies.
Seyyed (2012) provides further explanation that audit quality could be a function of the auditor's ability to detect material misstatements and reporting the errors. Together with other similar definitions, they all emphasize on two of the most important aspects of audit quality, namely auditor ability or auditor effort, and auditor independence. Therefore, this stream of definitions is mainly about the auditors' quality.
1. Audit independence Okolie and Izedonmi (2014) defined audit independence as an auditor's unbiasedness in taking decisions during an audit. Independence implies being free from inspiration, stimulus or guidance of which in the absence of independence, the value of the audit function will be greatly compromised. Prior studies suggest that high audit fees paid by a company to its external auditor enhances the economic ties between them and as such may compromise the independence of the auditor (Li and Lin, 2005). This weakened unconventionality results in failed audit quality and gives room for earnings manipulations (Okolie and Izedonmi, 2014). In past studies, audit fees have been used to measure auditors' independence (Wooten, 2003). An auditor's lack of independence increases the possibility of being perceived as not being objective. This means that the auditor will not likely report a discovered breach. Prior studies contend that high fees paid by a company to its external auditor increase the economic bond between the auditor and the client and thus the fees may impair the auditor's independence (Li & Lin, 2005). The impaired independence results in poor audit quality and allows for greater EM (resulting in lower earnings quality and stock estimation).
2. Audit Firm Size DeAngelo (1981) theorizes that larger firms perform better audits because they have a greater reputation at stake. In addition, because larger firms have more resources at their disposal, they can attract more highly skilled employees. Others have theorized that large auditors attract a fee premium because their greater wealth reduces clients' exposures in litigation. Others have theorized that there is no real audit quality difference, but the perception exists because large firms are well known and have gained a reputation for high quality. On the whole, the evidence is mixed, but it appears that there is some relationship between audit firm size and audit quality. Based on DeAngelo's (1981) reports, many other studies use auditor size (specifically Big8, Big6, Big5 or Big4 Vs non-Big8, non-Big6, non-Big5 or non-Big4) to differentiate audit quality levels (Krishnan, 2003).
3. Audit Firm Specialization Auditor industry specialization is another attribute of audit quality which extant literature suggests could affect earnings management practices of firms. This is because industry specialist auditors are familiar with the business operations of the industry of their specialization and also possess industry relevant experience and knowledge that enables them to audit companies in the industry more effectively than their counterparts (Minutti-Meza, 2013;Sarwoko & Agoes, 2014). The empirical link between auditor industry specialization and earnings management is documented by several studies, with more current ones contained in Jaggi, Gul and Lau (2012).
Research Journal of Finance and Accounting www.iiste.org ISSN 2222-1697 (Paper) ISSN 2222-2847 (Online) Vol.10, No.24, 2019 4. Auditor Tenure Auditor tenure is viewed as the length of time between auditor-client relationship (Okolie, 2014). A lengthy link between the auditor and his client may threaten unconventionality given developed familiarity. This may lead to less caution and compromise on the part of the auditor. Besides, a lengthy engagement may bring about less effort to signal the failings of internal control and risk sources (Okolie, 2014). Knapp (1991) established a linkage between audit tenure and competence. The objectiveness of an auditor in detecting anomalies increases in the first years of engagement but wanes with time, reaching its weakest level after 20 years of service (Okolie, 2014). There has been considerable decrease in number of years for auditor tenure. In the US, auditor tenure was reduced from seven to five years; the European Commission recommended a rotation of engagement partners every seven years; in France, auditors are chosen for six financial years, while in Nigeria audit engagement should not exceed three years (Okolie, 2014). For this study, auditor tenure will be proxied as length of auditorclient relationship using a dichotomous variable of "1" if 3 years+ and "0", if otherwise.

2.2.2
Earnings Management The accounting literature is yet to provide a universally accepted definition of the term "Earnings Management (EM)". Earnings management arises when managers exercise discretion over accounting numbers, with or without restrictions. The discretion is exercised to either maximize firm value (shareholder's wealth) or to maximize the selfish interest of managers (opportunistic earnings management). Most of the literature on EM deals with rounding earnings to reach cognitive reference points in income (EPS). Brenner and Brenner (1982) indicate that humans possess only a limited amount of memory and, therefore, store only the most relevant information about a number in their minds. Thus, investors would tend to remember primarily the most important digits within a number. For example, Carslaw (1988) suggests that investors would remember N398 million in earnings by rounding it down to N390 million or perhaps even to N300 million. Therefore, to prevent investors from rounding earnings down when unmanipulated income falls only slightly below user reference points, management will find ways to manipulate earnings to round it up to just across the breakpoint. In the example above, management would round earnings up from N398 million to something slightly above N400 million.
Significant research across many time periods and countries shows that this form of EM occurs with great regularity (Jordan et al., 2008). All these studies examine the second earnings digit and indicate that this digital position contains abnormally lower rates of high numbers (e.g., eights and nines) and higher frequencies of small numbers (e.g., zeros and ones) than would be anticipated in unmanipulated earnings. All other numbers appear in the second earnings position at rates approximating their expected frequencies. From this result, each researcher concludes similarly that when the second earnings digit falls just below zero, management rounds income up so that the second digit just reaches or crosses over zero. The object of this manipulation is to increase the most important digit (i.e., the first digit) by one. Thomas (2009) analyzes distributions of numbers in the right-most EPS digit (i.e., the digit two places right of the decimal point). He finds that for entities with positive earnings zeros and fives appear in this digital position far more often than expected. Das and Zhang (2003) examined the digital frequencies appearing in the position immediately right of the decimal point in the EPS numbers for a sample of companies. They conclude that their "evidence is consistent with managers manipulating earnings upward to round-up reported EPS." They further surmise that managers are more motivated to manipulate earnings to round up EPS when unmanipulated EPS numbers are close to behavioral benchmarks.
All three of the above studies examining digital patterns in EPS provide evidence of CEM in relation to EPS. More specifically, the research suggests that earnings are managed so that EPS can be rounded up to desirable reference points. Earnings Per Share (EPS) indicates the amount of net profit after tax (but before taking account of extra-ordinary income and expenses) attributable to each ordinary share in issue and ranking for dividend during the period. Shareholders are interested in this ratio because it indicates the amount of earning that will eventually be available to pay their dividends.

2.2.3
International Financial Reporting Standards effects on Earnings Management Numerous studies have investigated the impact of IFRS adoption on earnings management but there is no general consensus on the kind of impact it has. Cai et al. (2008) reported a declining earnings management after the voluntary and obligatory IFRS adoption from his study of 32 countries from 2000 to 2006. Barth et al. (2008) who studied 21 countries reported decreasing earnings management, more value relevance of accounting information and timely loss recognition during the post IFRS adoption period. This result was also supported by Christensen, Lee and Walker (2008), who submit that earnings management declined and timely loss recognition increased after IFRS adoption in Sweden and Germany. Thus, their findings stipulate that companies that adopt a higher quality international financial reporting framework would eventually have a higher financial reporting quality.
Ashbaugh, LaFond and Mayhew (2003) who examined 17 European nations identified decreasing earnings management and they maintained that IFRS requires higher information disclosure which enhances manipulation  (2015) investigated the impact of IFRS adoption on earnings management behaviours of listed deposit money banks in Nigeria. They reported that the restriction to incur losses under IFRS substantially reduces the banks' ability to engage in earnings management.

Empirical Review
Okolie, Izedonmi, and Enofe (2013) examined the effect of audit quality on earnings management of companies listed on the NSE for the period 2006-2011. Using a sample of 57 manufacturing firms and regression analysis to analyze the data, the study documented a significant negative association between audit quality (proxy by audit firm size, audit tenure, audit fees, audit client importance) and discretionary accruals of the sampled companies. Okolie (2014) investigated the relationship between audit quality and accrual-based earnings management of Nigerian firms. Using a sample of 57 non-financial firms listed on the NSE and the modified Jones model to measure discretionary accruals, the study documented a significant positive association between audit fees and discretionary accruals, and a negative association between audit tenure and discretionary accruals of Nigerian companies. Though the two Nigerian studies produced consistent results, there are also limited because all of them ignore the internal control aspect of audit quality (audit committee). Also, since the studies were not directly in relation to oil marketing companies in Nigeria, the findings are not likely to apply to oil companies in Nigeria due to sectoral differences between the sampled companies and oil companies in Nigeria. Aliyu, Musa and Zachariah (2015) examined the effect of audit quality (represented by audit firm size, joint audit and auditor financial dependence -a measure of client importance) on earnings management of listed deposit money banks in Nigeria. Earnings management proxy by discretionary loan loss provision was estimated using Beaver and Engel (1996) model, tested by researchers such as Fiechter and Meyer (2011). The study used a sample of seven (7) deposit money banks listed on the NSE for the period of 2006 -2013 while data analysis was done using ordinary least square (OLS) regression technique. Findings from data analysis indicated that both audit firm size and joint audit have significant negative effect on earnings management of listed deposit money banks in Nigeria. Auditor financial dependence had a significant positive effect on earnings management of listed deposit money banks in Nigeria during the study period. Though this study is from Nigeria, it is limited because given the peculiar nature of the banking sector; it is likely that its earnings management will be different from other sectors such as oil marketing companies not captured in the study sample. Consequently, some findings of the study may likely not apply to listed oil marketing companies in Nigeria. Also, findings from the study could be different from other sectors in Nigeria owing to the difference in the method of estimating discretionary accruals used for financial companies.
Recently, Tyokoso and Tsegba (2015) investigated the effect of audit quality on earnings management of listed oil marketing companies in Nigeria for the period 2004-2013. The dependent variable earnings management represented by Discretionary Accruals (DA) was estimated using the modified Jones model while the independent variable audit quality was represented by audit firm size, auditor industry specialization and auditor tenure. The findings of the study indicated that both audit firm size and auditor industry specialization have insignificant negative effect on DA of the sampled listed oil marketing companies in Nigeria during the period of study. In contrast, auditor tenure had a significant negative effect on DA of the companies. Umoren and Enang, (2015) empirically examines whether the mandatory adoption of IFRS has improved the value relevance of financial information in financial statements of commercial banks in Nigeria. The sample comprises of twelve listed banks in Nigeria. Specifically, financial statement figures of 2010 and 2011 (preadoption period) and 2012 and 2013 (post-adoption) were utilized. Descriptive statistics and least square regression were conducted to analyse the effect of IFRS adoption on the accounting quality. The result indicates that the equity value and earnings of banks are relatively value relevant to share prices under IFRS than under the previous Nigerian SAS. Results also indicate that earnings per share is incrementally value relevant during post-IFRS period while book value of equity per share is incrementally less value relevant during the post-IFRS period.
Yahaya, Yusuf & Dania, (2015) studied post-IFRS adoption value relevance of accounting information using two models. First, a price model which used proxies such as market price per share, book value of equity per share, earnings per share and cash flow per share. Second, a return model which used proxies such as annual return, earning per share, change in earning per share and the results showed that the explanatory power R2 for the price model specification is 84% for the total sample and that all coefficients are statistically significant. A comparison of coefficients indicates that the EPS of 3.47 has a higher explanatory power than any other variables. The results also demonstrate that explanatory power of accounting numbers increased from preadoption (60%) to post-adoption (78%). Similarly, Explanatory power (R2) for the return model specification is 13.4% for the total sample and just coefficient of EPS level is statistically significant. The explanatory power for the return model increased from pre-adoption (15.6%) to post-adoption (16.4%). According to both sub-samples just a coefficient of EPS level is statistically significant. So, the result of the return model also indicates adoption of IFRS improved relevance of accounting numbers in the deposit money banking sector. In view of these results, there is need for further study to explore the reasons for the superiority of EPS over BVEPS.

RESEARCH METHODOLOGY 3.1 Design, Sample and Data
The ex-post facto research design is adopted for this study based on positivist approach. An ex-post facto research design is used to describe the statistical effect of one variable on another. It is most appropriate for this study because it allows for testing of expected effects between audit quality proxies and shareholders' earnings of listed manufacturing companies in Nigeria and the making of predictions regarding such effects. The data for this study is obtained from secondary source. Secondary data were extracted from the published audited annual reports and accounts of the companies.
The population of the study comprises of all twenty-one (21) listed manufacturing firms on the Nigerian Stock Exchange (NSE) as at 31st December 2018 that are engaged in manufacturing activities. Having satisfied all the criteria used in filtering, eleven (11) firms serve as the sample of the study.
i. Shareholders' earnings measured as earnings per share (EPS). EPS is net profit after tax (but before taking account of extra-ordinary income and expenses) divided by ordinary share in issue. ii. Auditors' Independence (ADI) is measured as auditors' fees divided by firm revenue. iii. Audit firm size (AFS) is measured as "1" if the audit firm is in the Big-4, and "0" if otherwise. iv. Auditor tenure (ADT) is proxied as length of auditor-client relationship using a dummy variable of "1" if 3 years+ and "0", if otherwise. v. Audit firm specialization (ADS) is measured as "1" if the audit firm is industry specialist familiar with the business operations and industry possess knowledge for effective audit, and "0" if otherwise. It is the final proxy of audit quality. vi. International Financial Reporting Standards (IFRS) is a dummy variable, IFRS =1 for period between 2013-2018 (Post-IFRS), and IFRS=0 for period between 2007-2012 (Pre-IFRS). The empirical predictions developed in this section is derive from the properties of shareholder's earnings, market price of stock and proxies of audit quality employed in the study. Therefore, the empirical analysis begins by providing descriptive statistics relating to these components. Table 1 provides statistics on the characteristics of average, median, maximum, minimum and standard deviation of the variables included in the study.

Descriptive Statistics
The table indicated an average value of 2.41 for EPS. The minimum and maximum values of EPS during the study period are 11.85 and -6.37. These values imply that some sampled companies reported losses while others reported profits during the study period, with the average earning on shares being 214k.
Similarly, the table shows that auditor independence (ADI) has a mean value of 0.083 during the study www.iiste.org ISSN 2222-1697 (Paper) ISSN 2222-2847 (Online) Vol.10, No.24, 2019 period. This value implies that only about eight (8) percent of the sampled manufacturing companies during the period of the study were audited by auditors without perceived independence, indicating that about 92 percent of the firms are audited by perceived independent auditors. The minimum and maximum values of auditor independence stand 0 percent and 61.9 percent respectively. Overall, the values indicate that most sampled firms have independent external auditors auditing their financial reports.
The table further revealed an average value of 0.71 for audit firm size (AFS). The value implies that about seventy-one (71) percent of the sampled manufacturing companies were audited by the big 4 audit firms in Nigeria (KPMG, PWC, Ernst and Young Delloitte) during the study period. The mean value of seventy percent further suggests that only 29 percent of the sampled companies were audited by non-big 4 audit firms in Nigeria during the period of investigation. This shows that the audit market in the sector is dominated by the big 4 audit firms in Nigeria and just a few non-big 4 audit firms audited these manufacturing companies in Nigeria. The minimum and maximum values of audit firm size during the study period were zero (0) and one (1) respectively. The minimum and maximum values of audit firm size indicate that auditor size is measured by a dummy variable which takes the value of one if the company is audited by a big 4 audit firm and zero if otherwise.
The descriptive results for audit tenure (ADT) revealed an average value of 0.358. The value implies that about thirty-six (36) percent of the sampled companies had auditors who audited them for more than three (3) years during the study period, while about sixty-four (64) percent of the firms changed auditors within three years. The minimum and maximum values of audit tenure during the study period were zero (0) and one (1) respectively. The minimum and maximum values of audit tenure indicate that audit tenure is measured by a dummy variable which takes the value of one if the length of auditor-client relationship is 3 years+, and "0", if otherwise.
Data on audit firm specialization (ADS) reveals an average of 0.89, indicating that about 89 percent of audited firms had auditors who were industry specific auditors, understanding the economic conditions and operations specific to the manufacturing industry. About 11% of the firm-year sample recorded auditors who were not industry specialists. This result is also relevant to the study in analysing the effect of using the audit services of audits who are specialists in the manufacturing industry.  Table 2 presents correlation values between dependent and independent variables and the correlation among the independent variables themselves. These values are generated from Pearson Correlation output.
Auditor independence is positively correlated with EPS (0.705) of listed manufacturing companies in Nigeria during the study period. The high positive coefficient between auditor independence and EPS of the sampled firms in Nigeria suggests that auditor independence is associated with improved and reliable EPS figure of firms.
The table also revealed a positive correlation coefficient between audit firm specialization and EPS (0.2003) of the sampled manufacturing firms in Nigeria during the period of investigation. The significant positive coefficient between audit firm specialization and EPS of the sampled companies is an indication that auditors who are industry specialists are associated with reliable EPS figure of listed manufacturing companies during the study period. This relationship is not surprising because having knowledge, expertise and experience about the economic condition and the operations of the industry, will lead to more effective industry specific audit, thus, enhancing reliable earnings practices, reporting and disclosure.
The table further revealed a positive correlation coefficient between audit firm size and EPS (0.2976) of the sampled firms in Nigeria during the period of investigation. The significant positive coefficient between audit firm size and EPS of the sampled companies is an indication that audit firm size is associated with reliable EPS figure of quoted manufacturing firms during the study period. This relationship is not surprising because big 4 audit firms have the resources and capacity to perform high quality audit that is capable of producing reliable reported earnings firms.
However, auditor tenure is negatively associated with the EPS figure of sampled firms (-0.0628). The negative relationship between auditor tenure and EPS of Manufacturing companies in Nigeria shows that auditor tenure is associated with decrease in EPS figure credibility. This relationship is expected because lengthy link between the auditor and his client may threaten unconventionality given developed familiarity. This may lead to less caution and compromise on the part of the auditor. Besides, a lengthy engagement may bring about less effort to signal the failings of internal control and risk sources.
Finally, it is observed that among the proxies of audit quality, audit independence is positively related with Research Journal of Finance and Accounting www.iiste.org ISSN 2222-1697 (Paper) ISSN 2222-2847(Online) Vol.10, No.24, 2019 audit firm specialization and audit firm size, implying that an independent audit review is associated with the industry specialty of the audit firm and how large the audit firm is (in terms of coverage). Table 3: Average Slope (coefficient), t-statistics and P-value from annual Cross section regressions of hypothesis 1 Audit independence (ADI) effect on EPS and MPS The regression result shows that ADI is a positive predictor of firm EPS. Stated differently, the more independent the auditors, the more reliable the EPS reported in the financial statements of Manufacturing firms in Nigeria. The effect is found to be about 13.78, implying that audit independence is regarded as audit quality, which allows for greater confidence of the market in the reported accounting numbers of firms. The t-statistics value of 8.436 and p-value of 0.000 less than the 0.05 alpha level shows that the effect is statistically significant.

4.3.2
Audit firm size (AFS) effect on EPS The regression result shows that AFS has a positive effect on EPS figures of Manufacturing firms listed in the NSE. Stated differently, the larger the audit firm size, the higher the dependence on reported EPS, revealed by a positive coefficient of 3.175 (p=0.03). This is because large audit firms have more resources to conduct high quality audits, have a large client base which makes them less dependent on any one client that could make them compromise their audit quality and also have more investment in reputation capital which is at stake if they are found to have compromised audit quality than small audit firms. This therefore enhances market confidence on reported EPS figures of manufacturing firms in Nigeria, Based on the result, audit firm size has significant effect on EPS of manufacturing companies in Nigeria.

4.3.3
Audit Tenure (ADT) effect on EPS The regression result shows that ADT is a negative yet insignificant predictor of the EPS figure of Manufacturing firms in Nigeria (coefficient = -0.628, p=0.3993). Stated differently, the longer the tenure of an audit firm, the less reliable the EPS figure reported in the financial statement, which results in plummeting of share prices. This result implies that long auditor tenure is associated with less EPS. The result is expected because as the length of auditor-client relationship increases, familiarity with the audited firm may cause compromise, which threatens the reliability of reported EPS. 4.3.4 Audit Firm Specialization effect on EPS and MPS Audit firm specialization has been found to affect EPS, with the effect being positive and statistically significant. The effect of ADS on EPS is given by a coefficient of 1.196 (with t-statistics=3.692 and p=0.019). The result is expected because Industry specialist auditors are familiar with the business operations of the industry of their specialization, and, also possess industry relevant experience and knowledge that enables them to audit companies in the industry more effectively than their counterparts. This makes the market rely more on the EPS figure published by the firms. The findings collaborate with the findings of Jaggi, Gul and Lau (2012). Based on the result, audit firm specialization has significant effect on EPS of Manufacturing companies in Nigeria.  Yahaya et al (2015) who found that IFRS adoption reduces the earnings management behaviours of bank firms in Nigeria as the restriction to incur losses under IFRS substantially reduces the banks' ability to engage in earnings management.
The difference between the impact of audit independence on EPS in the pre-and post IFRS periods is about 1.084 (i.e. 4.196 (for post IFRS effect) minus 3.112 (for pre-IFRS effect)). The difference is significantly significant at the 0.05 level, using the Wald coefficient restriction test. Thus, there is a significant difference in the effect of audit independence on EPS in the pre-and post-IFRS The difference between the impact of audit firm size on EPS in the pre-and post IFRS periods is about 1.126 (i.e. 3.112 (for post IFRS effect) minus 1.986 (for pre-IFRS effect)). The difference is significantly significant at the 0.05 level, using the Wald coefficient restriction test. The result holds that the effect of audit firm size in the post IFRS period is higher, and significantly different from the effect of audit firm size in the pre IFRS period.
The difference between the impact of audit tenure on EPS in the pre-and post IFRS periods is about not significant in the pre-and post-IFRS periods. The difference is not significant at the 0.05 level, using the Wald coefficient restriction test, indicating that the effect of audit firm tenure in the post IFRS period is not different from the effect of audit tenure in the pre IFRS period.
The difference between the impact of audit firm specialization on EPS in the pre-and post IFRS periods is about 1.869 (i.e. 4.214 (for post IFRS effect) minus 2.345 (for pre-IFRS effect)). The difference is significantly significant at the 0.05 level, using the Wald coefficient restriction test. The result holds that the effect of audit firm specialization in the post IFRS period is higher, and significantly different from the effect of audit firm specialization in the pre IFRS period.

Conclusion
The paper therefore reaches the following conclusions based on the stimulating result of data analysis and discussion: The study provided empirical evidence on the association between audit quality (proxied by auditors' independence, audit firm size, and auditor tenure) and shareholders' earnings (proxied by EPS) of listed Manufacturing companies in Nigeria. Specifically, the study concluded that auditors' independence, audit firm size and audit firm specialization have positive and significant effect on EPS of Manufacturing companies listed in the Nigeria Stock Exchange. This indicate that auditors' independence, audit firm size and audit firm specialization constrain earnings management of the sampled firms, resulting in quality, credible and reliable EPS figure reported by the sampled firms, thereby increasing the value of the firm. The correlation analysis also supports this finding, with positive significant correlation coefficients results found between the said variables.
Overall, audit quality has significant effect on shareholders' earnings and market price of stocks, among listed Manufacturing companies in Nigeria.
Based on the findings of this study, Public companies (and more specifically the Manufacturing firms) should ensure that auditors they hire are regarded as large, independent, and industry specialists. This will ensure that the public relies on the EPS figure of such firms, which can in turn can affect their share pricing. Thus, for improved market confidence and share valuation of manufacturing firms, the role audit quality cannot be overemphasized.
Again, auditors should be rotated regularly to diminish the negative effects of protracted audit tenure on the market pricing of stocks in Nigeria.
Finally, investors and stakeholders should pursue more future signaling disclosures, that provide information of all earnings quality with their implication for future stock performance.