Audit Assertions and Change of Auditors’ Opinion in the Brazilian Market

Received in 10/17/17. Ask to Revise on 12/27/17. Resubmitted on 2/8/18. Accepted on 2/26/18 by Dr. Gerlando Augusto Sampaio Franco de Lima (Assistant Editor) and by Dr. Orleans Silva Martins (Editor). Published on 3/29/2018. Organization responsible for the journal: Abracicon Abstract Objective: Investigate the use of audit assertions to justify opinion changes in Brazilian audit reports. Method: A descriptive study was undertaken through content analysis, in which 2,243 reports from 338 non-financial publicly traded companies listed on BM&FBOVESPA between 2009 and 2015 were investigated, identifying 192 audit reports with a changed opinion. We investigated whether some assertion prevails in the opinion changes and whether each of them can be associated with certain accounting equity and income groups. Results: The audit assertions Evaluation and Integrity are the most used to justify opinion changes. As for the association between the audit assertions and the accounting groups, it was verified that the categories of Existence/Occurrence tend to be associated with asset and income accounts, while the Integrity categories are related to liabilities and expenses. The relevant number of justifications based on business continuity aspects stood out. Contributions: This study contributes to the development of the Brazilian literature on auditing and to reflections on the quality of audit work, mainly by presenting evidence on how and in what dimension the Brazilian auditors use audit assertions to justify opinion changes on the financial statements.


Introduction
Considering its purpose of contributing to an economic environment characterized by the credibility and reliability of financial information, reducing the informational asymmetries between the management and financial statement users (Dantas, Chaves, Silva & Carvalho, 2011), the audit consists of a systematic process to obtain and evaluate evidence on management assertions about economic actions and events to assess the degree of correspondence among them and the criteria set forth in the applicable financial reporting framework in order to express an opinion on whether or not the financial statements present misstatements. This definition, presented by the American Accounting Association (AAA, 1972), has reflected a certain scholarly and professional consensus on the scope and purposes of the audit, being reproduced with variations in textbooks - Boynton, Johnson and Kell (2002), Ricchiute (2002), Hayes, Dassen, Schilder and Wallage (2005) and Gramling, Rittenberg and Johnstone (2012), for example -and supporting professional standards -in particular the International Auditing Standards (ISA), which in Brazil are embodied in the Brazilian Accounting Standards of Independent Audit (NBC-TA).
As noticed, the formation of the auditor's opinion presupposes knowledge on the audited company, the environment it operates in and the nature of its operations. This knowledge provides the basis for the auditor to create expectations about the audit assertions in the financial statements, about which the auditor should obtain and evaluate evidence that confirms them or not (Felix & Kinney, 1982).
A model initially developed by Mautz andSharaf (1961, apud Leslie, Aldersley, Cockbum, &Reiter, 1986), the audit assertions, despite representing the point of reference that guides the development of audit work, has not been studied frequently in academic research. Internationally, Leslie et al. (1986), Smieliauskas and Smith (1990) and Waller (1993) can be cited. In the Brazilian accounting literature, no articles specifically discussing the topic were found.
In the mid-1980's, Leslie et al. (1986) evaluated this lack of studies from the perspective that researchers would be afraid to address a topic that would be at the forefront of practice because audit firms were still unable to integrate audit assertions into their audit approaches. Three decades later, even with the consolidation of the model, including in the professional standards, there remains a certain academic silence about the subject. In this case, exactly the opposite argument can be predicted, that is, with the wide use of the model in practical terms, the researchers may feel little stimulated to study the audit assertions. As Hartmann (2017) points out, however, it is important that accounting research also includes discussions on practical issues, with a view to enhancing its development by overcoming the barriers between theory and practice.
In this sense, understanding how auditors use audit assertions for targeting, applying procedures and obtaining appropriate and sufficient evidence, in order to anchor the formation of opinion on the financial statements, may be important for qualitative research in the area, especially as regards the quality of audit work, as Smieliauskas and Smith (1990) point out, by stating that audit quality can be improved and achieved through appropriate explanations linking audit evidence to audit assertions.
The problem is that the documentation of the audit process is, by legal and regulatory definition, confidential, making it difficult to carry out studies with such a degree of comprehensiveness -the only part of the audit work that is known to external users is the called the auditor's report, published with the financial statements. Thus, the researcher can only examine the application of the audit assertions in cases where these are explained in the auditor's reports.
In that context, the purpose of this study was to investigate the use of audit assertions by Brazilian auditors in order to justify the changes of opinion in the audit reports. In such cases, the auditor should justify the reasons why the statements present material misstatements, assuming they are related to the audit assertions. In more specific terms, we try to understand: if some audit assertion is prevalent to justify the change of opinion in the Brazilian market; and whether each audit assertion can be associated with asset or liability and income statement accounts -revenues or expenses.

Audit Assertions and Change of Auditors' Opinion in the Brazilian Market
In order to achieve the proposed objective, the basic paragraphs for the opinion in 192 audit reports with changed opinions were investigated, previously selected in a set of 2,243 reports on the annual financial statements of 338 non-financial publicly-traded companies listed on BM&FBOVESPA between 2009 and 2015. Besides this introduction, which contextualizes the theme and sets its objectives, this study addresses: the theoretical framework to support the understanding of fundamental issues, with a review of the audit assertions and a discussion about prior studies on the theme (Section 2); the methodological procedures adopted for the empirical tests (Section 3); the presentation and analysis of the results (Section 4); and, finally, the final considerations on the study (Section 5).

Audit assertions as a framework for audit work
ISA 300 (in Brazil, the ISAs were fully received, through the NBC-TA, with equivalent numbering, which is why, whenever reference is made to a particular ISA in this study, an NBC-TA exists with the same number and equivalent content) states that the objective of the auditor is to plan the audit to be performed effectively, with the planning corresponding to the definition of an overall strategy for the work and for the development of an audit plan. The Public Company Accounting Oversight Board (PCAOB, 2010) states that obtaining an understanding of the entity and its environment, as well as its internal control, is an essential part of planning an audit to respond to the assessment of the risks of material misstatements.
In this planning process, it is of particular importance to consider that the auditor, in declaring that the financial statements are in conformity with the applicable financial reporting framework, took into account that management makes implicit or explicit assertions regarding the recognition, measurement, presentation and disclosure of the various elements of these statements (ISA 315). The standard establishes that the auditor should use these statements to consider the different types of potential misstatements that may occur in the entity. This specification helps in the planning of audit tests and the auditor should test these assertions by obtaining appropriate and sufficient audit evidence (Gramling, Rittenberg, & Johnstone, 2012).
Although ISA 315 provides for audit assertions about classes of transactions and events, balance of accounts and presentation and disclosure, the standard itself admits that the auditor may treat these statements differently by combining, for example, statements about transactions and events with those related to the balance of accounts. In this sense, based on Hayes, Dassen, Schilder and Wallage (2005) and Gramling, Rittenberg and Johnstone (2012), it can be concluded that the most common has been the use of five audit assertions, as summarized in Table 1. Table 1 Audit Assertions

Existence / Occurrence
Refers to the existence of everything that is registered in the assets, liabilities, equity and income accounts.

Integrity / Completeness
Its meaning is opposite to existence/occurrence, that is, the equity and income items should have been recorded completely -in short, everything that exists or occurred should have been recorded.

Rights and Obligations
Intends to verify if the organization controls or is entitled to its assets and if its obligations (liabilities) are truly rights of third parties.

Valuation / Allocation
Is associated with how the entity assessed its equity and income items. They should have been valued and adjusted in accordance with the accounting practices concerning adjustment at realizable value, depreciation calculations, price-level restatement calculation, among others.

Presentation / Disclosure
Refers to the disclosure of relevant information in the financial statements, particularly in the notes to the financial statements, complying with the accounting standards and guaranteeing that the transactions are clarified to the users.

De Mautz and Sharaf (1961) to ISAs: evolution and consolidation of audit assertions
According to Leslie et al. (1986), the first reference to the concept of audit assertions in audit literature was found in Mautz and Sharaf 's The Philosophy of Auditing, written in 1961, and remained hibernating until 1973, when it reappeared in A Statement of Basic Auditing Concept (Asobac), a publication on basic audit concepts. Still in the 1970s, R.J. Anderson also recognized the merits of the concept and was responsible for the recognition of audit assertions in the manual of the Canadian Institute of Chartered Accountants (CICA), today called the Auditing and Assurance Standards Board (AASB). It was only in 1980 that the American Institute of CPAs (AICPA) addressed the concept for the first time through SAS 31.
Nevertheless, not many references to the audit assertions are found in the accounting literature. In the 1980s, Leslie et al. (1986) attributed this lack of studies in the field to the fact that audit firms were unable to integrate the concept of assertions into their audit approach. Thus, academics might be reluctant to produce studies that would be one step ahead of practice and would rather study what they believe to be the most commonly used method.
In the early 1990s, Smieliauskas and Smith (1990) stated that the relationship between detailed audit procedures and elementary assertions is not yet standardized in practice, which can be caused by differences in expertise, cost structure and cost/benefit of audit firms.
More recently, the concept of audit assertions is already consolidated and addressed in the International Standards on Auditing (ISA). Thus, the auditors need to prepare their audit work based on the assertions, in order to test them in the course of the work, as explained in Section 2.1.

Theoretical discussions on the utility of audit assertions
Even though the issue has been discussed since 1961 and consolidated as regulatory professional practice, the academic studies on audit assertions are still not frequent, even in the international scenario, with the work of Leslie et al. (1986), Smieliauskas and Smith (1990) and Waller (1993). Leslie et al. (1986) set out an audit approach based on audit assertions, focusing on the strategic level and seeking to refine the internal control review and evaluation approach conditioned to the inherent risk assessment, to guide the auditor to a more comprehensive view of the financial statements, join the different audit concepts and use the audit assertions as the organizing principle of these concepts, proposing an audit method that would truly meet the precepts of practicality and effectiveness.
In their study, Smieliauskas and Smith (1990) refined the theory of Audit Evidence based on research in philosophy of science. Based on this precept, they improved the definition of evidence confirmation and related it to the auditor's opinions and found that explanations are a critical component linking the evidence to audit assertions. Finally, they concluded that the definition of professional audit standards should increasingly emphasize the underlying explanations of audit evaluations. Smieliauskas and Smith (1990) also reinforce that purely quantitative audit models may not be sufficient to develop a theory about audit evidence. The explanatory component of the evidence helps to ensure that the professional standards and the auditor's social duty to bring confidence to external users are met and also provide an important factor associated with the quality of the audit. The authors demonstrated that appropriate explanations improve the quality of audit work.
Therefore, it is expected that problems in the statements will lead to a modification in the auditor's opinion. Audit assertions provide coverage for the main sources of errors in the financial statements (Smieliauskas & Smith, 1990). Thus, obtaining evidence as to the presence or not of material misstatements related to these categories of assertion influences the opinion that the auditor should issue. In addition, an opinion based on these assertions helps to construct a sound and high-quality explanation for the audit report.
In addition to Leslie et al. (1986) and Smieliauskas and Smith (1990), another author on the subject was Waller (1993). Although not directly addressing the issue, he studied the association between inherent and control risks and statements. It was expected that inherent risk and control assessments would vary according to the assertions of each account, which was not confirmed.
Although few, international studies were able to address the audit assertions more specifically. The Brazilian literature does not present studies that deal directly with the subject, only studies on other aspects of the auditor's opinion, as presented in Section 2.4.

Research in Brazil on opinion change
Although no studies were found in the Brazilian literature that specifically address audit assertions, studies on changes in the auditor's opinion are relatively common, focusing either on their causes or their consequences, among which we can mention Batista, Pereira, Silva and Imoniana (2010), Damascena, Firmino andPaulo (2011), Camargo (2012), and Dantas, Barreto and Carvalho (2017).
According to ISA 705, the modified opinion should be issued when audit evidence shows that the financial statements present material misstatements or when the auditor is unable to obtain appropriate and sufficient audit evidence to express an opinion on the existence, or not, of relevant misstatements, taking three forms: qualified opinion, used when there are relevant misstatements or when it was not possible to obtain appropriate and sufficient audit evidence; adverse opinion and no opinion, used when there are widespread effects, the former being used if relevant misstatements exist and the second when there is no possibility of obtaining appropriate and sufficient evidence.
The study by Batista et al. (2010) sought to analyze the reaction of the stock returns to the disclosure of audit opinions. Through a bibliographic and documentary research and secondary data collection, the authors identified whether the opinions were issued with or without qualification and obtained the average stock returns, which permitted the analysis of the data using the Wilcoxon test, thus seeking to compare the average stock return before and after the disclosure of the audit opinion. They concluded that the publication of the opinions did not influence the average stock returns in the month after their disclosure. Damascena et al. (2011) aimed to identify the factors that motivate the issuing of a qualified opinion and/or paragraphs of emphasis in the audit reports on the Brazilian public companies' statements. In total, 1,466 financial statements were analyzed from 2006 to 2008, 647 of which presented an opinion with qualifications and/or paragraphs of emphasis. After analyzing the content, the authors concluded that the limitation in the scope and impossibility of opinion formation are the reasons that most lead to the issuing of qualified opinions and that continuous losses, short-term liabilities and working capital shortages lead to the existence of paragraphs of emphasis. Camargo (2012) investigated the determinants of opinion in the audit reports of 279 companies listed on BM & FBOVESPA in 2010, concluding that companies with greater delays in receiving opinions and those audited by the big four are more prone to changed opinions, while companies that paid the highest fees and exchanged firm in the analyzed period are more likely to receive unchanged opinions. Dantas et. al. (2017), in turn, assessed the impacts of the changed opinion on the continuity of the audit service contract and whether the fact that the audit firm is a big four or that the audited company belongs to the corporate governance segments reduce this risk of contractual discontinuity. We analyzed 333 companies listed on BM & FBOVESPA, considering the period from 2009 to 2014 and the authors concluded that there is a positive and statistically relevant relationship between the issuing of a modified opinion in one period and the change of auditors in the subsequent period. They also found that the audit firm being one of the big four or the company's listing in a corporate governance segment of the stock exchange does not reduce the risk of discontinuity of the contracts.

Thayanne Costa da Silva, José Alves Dantas
In summary, although the studies carried out in Brazil related to the auditor's opinion, particularly the context in which reports with changed opinions were issued, none of them addresses the issue from the perspective of audit assertions, defined as guides for audit work, which makes this study a pioneer in the Brazilian literature.

Methodological Procedures
Considering its objectives, this is a descriptive study with a qualitative approach. Prodanov and Freitas (2013) define descriptive research as studies that expose the characteristics of a certain population or phenomenon and define the qualitative approach as using the natural environment as a source to collect data, interpret phenomena and attribute meanings.

Study population
In view of the objectives and exploratory nature of the study, first, we examined the audit reports on the annual financial statements of 338 non-financial publicly traded companies listed on BM&FBOVES-PA between 2009 and 2015, totaling 2,243 reports, directly collected on the website of the Brazilian Securities Commission (CVM).
As the study is focused on the audit reports with opinion changes, we concentrate on identifying the type of opinion the auditors issued in the previously identified reports, whose statistics are summarized in Table 2. The results of this preliminary assessment reveal that, in the set of reports examined, 192 (9% of the total) contained changed opinions -with qualification or no opinion -which were the object of this study.
Audit Assertions and Change of Auditors' Opinion in the Brazilian Market

Analysis criteria
Based on the 192 modified opinion reports specified in Section 3.1, the basic paragraphs for the opinion were analyzed by means of content analysis, aiming to identify the audit assertions used as justification for the change of opinion, using the content of Table 1 for reference.
For a better understanding of how audit assertions are used to justify auditors' opinion changes, the data in this content analysis is organized in statistical terms using the following criteria: • Relation between the audit assertions and the type of report issued, in order to evaluate if any assertion category is prevalent to justify the opinion change; and • Relationship between the audit assertions and the equity and profit and loss account groupsassets, liabilities, shareholders' equity, income and expenses -in order to identify whether it is possible to associate each category of statement with a particular accounting group.
It should be noted that the process of classifying the justifications for opinion changes, based on the content analysis of the report, is an intrinsically subjective task, subject to the bias of the researcher's judgment, which obviously characterizes a limitation of this study.

Analysis of Results
As established in Section 3, the baseline paragraphs for the opinion expressed in the 192 reports with changed opinions that constituted the study population were investigated, aiming to identify the justifications used for this change, as well as the association between these reasons and the audit assertions. The purpose is to assess whether any audit assertion prevails to support the opinion change, and whether each of them can be associated with specific groups of equity and income accounts.

Relation between audit assertions and the type of report issued
The first step in the analysis of the results was to relate the audit assertions with the types of changed opinion, with a view to concluding whether any assertion category stands out among the justifications for opinion changes and whether the types of reports vary. The results have been consolidated in Table 3. Some reports presented more than one justification for opinion changes, which required the allocation in more than one category, resulting in 360 justifications extracted from the 192 reports with changed opinions studied here. Although referring to the categories presented in Table 1, a special topic was needed to frame the issues addressed in the basic paragraphs for the changed opinion, related to Business Continuity -which includes justifications that refer to the risk that the disclosing entity will not continue to carry out its operations, that is, which discusses evidence that questions the assumption that the organization will continue to operate normally. Justifications for opinion changes that could not be classified in the other categories were classified as "other" due to the difficult association with some of the audit assertions or risk of operational continuity.
In relation to the audit assertions, the results initially show that the categories Evaluation and Integrity are the most frequent arguments for opinion changes in all the reports studied, both with a frequency of 23%. This reveals that the evidence associated with non-recognition of equity or income items or the misstatement of these items are the matters the auditors address most to justify their giving a changed opinion on the statements. These results arouse a question as to whether the concentration in these two categories is due to the auditors' greater concern with such matters or the fact that the companies commit more improprieties in these areas. Considering that the only visible part of the auditor's work is his report, that question cannot be answered objectively.
For the total group of examined reports, the categories Evaluation and Integrity have the same frequency. When considering the type of opinion, a clear distinction is noted regarding the prevalence of audit assertions. For the reports with qualifications, the category Evaluation (29%) is the most used for justification, while Integrity (26%) prevails among the arguments for the auditor to issue a report without opinion. It should be noted, however, that operational continuity problems account for 31% of the justifications used in reports with no opinion, suggesting that, when the client shows signs of discontinuity, the auditor is more afraid to express an opinion. This set of evidence suggests that problems of misstatements in the value of equity items and/or results justify exemptions, while problems of non-recognition of these items and operational continuity result in a lack of opinion. This is consistent with Serra and Rodríguez' (2012) statement that the auditor, when issuing reports with opinion changes, risks losing clients. On the other hand, when expressing a clear opinion when the situation requires modification, they face reputational losses and can even be held civilly liable to shareholders and investors who have been harmed or misled based on the content of the report. Thus, in delicate situations, such as cases where evidence, even if not conclusive, of a risk of discontinuity is identified, failure to issue a changed opinion would be too risky for the auditor, which may explain the apparently surprising number of cases in which operational continuity is addressed as an argument to question the adequacy of the financial statements.
At the other extreme, audit assertions related to Rights and Obligations (5%) and Presentation / Dissemination (6%) are the least used by the auditors to justify the change of opinion, which may suggest that they find problems of misstatements linked to these categories less important compared to the others. Nevertheless, the possibility of less space for manipulation regarding these matters or even less incentives for the management to stop complying with the regulatory determinations cannot be discarded.

Relation between audit assertions and the account groups covered by the change
After completing the analysis of the association between audit assertions and types of opinion change, the next step was to relate the assertion categories to the account groups that were subject to change, that is, that presented material misstatements. The purpose is to identify whether each audit assertion can be associated to a particular account group.
The first category is Existence/Occurrence and is related to the assertion that the equity items recorded in the statements actually exist and that the events translated into the income and expense accounts actually occurred. There were 40 references to this audit assertion in the 192 reports examined with changed opinion. The distribution of these cases by account groups has been consolidated in Table 4. As evidenced, 95% of the cases of opinion change based on the Existence/Occurrence category are linked to Asset or Revenue accounts, that is, they represent questions about the existence of assets disclosed in the Balance Sheet or the occurrence of revenues that are part of the income statement. The most recurring cases are related to financial assets and property, plant and equipment. These findings are consistent with the premise that management has more incentives to overestimate than to underestimate assets and revenues.
It is also consistent with Martinez' (2001) statement that a number of reasons can cause the manipulation of accounting income, including the motivations related to capital markets and contracts, with incentives to practice information management, aiming to reducing the investors' perception of company risks and increasing the managers' remuneration -based on profit. Therefore, the better the entity presents itself, in terms of equity position or performance, the better it will be for the manager. Hence the incentive to overestimate assets and revenues, which should be considered in the audit mechanisms to assess risks of material misstatements. The results shown in Table 4 are consistent with this premise.
In the case of the audit assertion Integrity, which refers to the questions about non-registration of existing obligations and expenses that occurred, 82 references were found in the 192 reports with changed opinions studied here. Of these, 90% are linked to liability or expense accounts, as evidenced in Table 5. This concentration in liability and expense accounts is also consistent with the management's incentives and, consequently, the audit risks. The cases of integrity problems identified in these accounts mainly concerned tax provisions and obligations. Following Martinez (2001), it is natural that management has more incentives to underestimate than to overestimate liabilities and expenses accounts in order to improve the economic-financial situation and performance reported through its statements. Furthermore, in this case, the findings highlighted in Table 5 confirm these theoretical perspectives.
The Rights / Obligations category is related to the premise of the management's assertion that the organization controls or holds rights over its assets and that its obligations are truly the rights of third parties. Only 17 references to this audit assertion were registered in the 192 reports with changed opinion and, in 82% of these citations, they could not be linked to a specific account group. The reason for not linking the large majority of problems to a particular account group is explained by the lack of more detailed explanations in the reports. The justifications related to Rights/Obligations were generally presented to indicate noncompliance with contractual clauses or problems, without further specification, which made it impossible to associate the problem with a specific account group. This limitation is associated with the fact that the auditor's report is the only part of the audit work that is accessible for review. Moreover, this lack of explanations also draws attention to the quality of the audit report made available to users, as this document is important in order to enhance the credibility of financial statements and assist in decision making. It is possible that, with the adoption of the new audit report, as from the base year 2016, some effect may exist in the way auditors communicate their findings.
For the audit assertion Valuation/Allocation, being associated with the valuation criteria of equity and income items the management adopts, notably with respect to accounting practices involving some degree of subjectivity and professional judgment, the prevalence of some account group does not appear to be a theoretical imperative. The consolidation, according to Table 7, of the 83 citations of problems related to the Valuation category to justify a change of opinion in the 192 reports examined, reveals a relatively significant concentration in the account group Assets though. Almost 60% of the cases related to the value of equity and income items that gave rise to changes in auditors' opinion are related to Assets accounts, mainly property, plant and equipment, presenting problems in the depreciation calculations. This seems to suggest, as opposed to what was initially anticipated, that there is more room for errors (or manipulations) in the measuring of asset items, although it was the audit assertion in which there were records of justifications for changes in the auditors' opinion related to all account groups.
In the case of the fifth audit assertion highlighted in Table 1, this is linked to the disclosure of the relevant information in the financial statements, particularly in the explanatory notes, in order to ensure the users' proper understanding of the economic and financial situation of the disclosing entity. As would be natural, given the greater concern with the production of explanatory notes, the analysis of the 22 cases of justifications for opinion changes based on this assertion category reveals that, in practically all of them (95%), there is no link to an account group, as evidenced in Table 8. Perc.
The topics in this category the auditors used to change their opinion on the financial statements as a whole include: non-disclosure of all required items in the consolidated financial statements; problems that jeopardize the company's operational continuity; and failure to present the financial statements in accordance with applicable laws and regulations. These are general matters, not associated with a specific account group. As for the only subject in the category related to Shareholders' Equity, it deals with problems in the restatement of adjustments of previous years. The Company recorded an adjustment of the equity method of accounting for the previous year in shareholders' equity, without restating the balances adjusted in accordance with the current standard, which caused an increase in equity.
Finally, it is worth noting that, although it is not specifically an audit assertion, which is the focus of this study, the distribution of cases of opinion change justified by risks of operational discontinuity was evaluated, given the relevance of the topic highlighted in Section 4.1. The 61 cases of this kind do not refer specifically to an account group, as evidenced in Table 9. Given the natural range the risk of operational discontinuity necessarily implies, the lack of reference to a specific account group is completely natural.

Final Considerations
A model initially developed in the 1960s, the audit assertions have turned into a relevant instrument for the planning, execution and formation of audit opinions, consolidating themselves as an integral element of professional standards. The assumption is that they help to identify possible material misstatements and should be used to justify a possible change in the auditor's opinion.
Notwithstanding their importance for audit work, the audit assertions have been little studied in the academic world, perhaps because they are already considered as something consolidated, in view of the regulatory determinations for their use. Considering, however, the argument that the study of practical issues may be important to enhance their development, this study aimed to investigate Brazilian auditors' use of these audit assertions to justify the changes of opinion in their reports, more specifically if some audit assertion prevails and if each of the assertions can be associated to equity and income accounts.
From a set of 2,243 reports on the financial statements of 338 non-financial publicly-traded companies listed on BM & FBOVESPA from 2009 to 2015, 192 audit reports with opinion changes were identified, which were studied in this research.
The results showed that, from a set of 360 justifications extracted from the analyzed reports, the assertions related to Evaluation and Integrity, both with a frequency of 23%, are the most recurrent to justify the opinion changes. The concentration in these assertions may be due to both a greater concern of the auditors with issues related to these categories and the fact that the companies commit more improprieties in these areas. Only the analysis of the audit report, which is the only visible part of the auditor's work, does not provide an objective answer to this question, which is a limitation of this study -due to the restricted access to the auditors' working papers.
At the other extreme, audit assertions related to Rights/Obligations and Presentation/Disclosure are the least used as a basis for opinion change, which may indicate a trend for auditors to assign less importance to problems related to these categories when compared to the others. In any case, one cannot overlook the fact that there may be less room for manipulation in relation to these assertions or even less incentive for management to disregard the regulatory instructions.
Regarding the types of opinion changes, there were differences regarding the prevalence of audit assertions. When it comes to qualified opinions, the most frequent assertion category is Evaluation (29%), while Integrity (26%) is the most used to justify cases of abstention. It is important to note that problems related to Operational Continuity, which does not represent an audit assertion per se, constitute 31% of the reasons for abstention, which may be related to the risk for the auditor's reputation in expressing an opinion in cases in which the companies present problems that can interrupt their activities.
In the second stage of the tests, the relationship between the audit assertions and the equity and income account groups was analyzed. The results showed that 95% of the cases of justification of opinion changes based on the Existence/Occurrence category are associated with asset and income accounts, while 90% of Integrity problems refer to liability and expense accounts. These results support the premise that management has more incentives to overestimate assets and revenues and to underestimate liabilities and expenses, which may justify the auditors' concern.
Also noteworthy is the fact that 58% of the problems reported as justifications for opinion changes related to the audit assertion Evaluation are related to assets, mainly to property, plant and equipment. This suggests the greater probability of material misstatements associated with the measuring of asset items, especially fixed assets, which can be explained by the subjectivity implicit in the disclosure of depreciation and impairment of these elements. Misstatements related to Rights/Obligations and Presentation/ Disclosure could not be associated with a specific account group.
This study contributes to the development of the Brazilian audit literature, mainly for presenting evidence on how and to what extent Brazilian auditors use audit assertions to justify a change of opinion on the financial statements examined. The audit assertions are expected to direct the audit work. The reflections and the study on the theme indicate that a high-quality audit seeks evidence based on the assertions present in the statements. Thus, it is interesting to note that the planning of the audit work may tend to value certain audit assertions, as presented in the results, which can raise doubts about the auditors' direction in their work and, consequently, doubts about the quality of the audit work.
As a contribution to the literature, it is also important to highlight the fact that this study permits an objective link between the theoretical foundations and their practical application in the process of planning, realization and communication of audit results.
Nevertheless, this study is limited not only by the restricted access to other audit information than the report that is part of the statements, but also by the subjectivity implicit in the content analysis method used in the analysis of the reports, being subject to the bias of the researcher's judgment.
As a suggestion for future research that complements the reflections and evidences highlighted in this study, we can highlight: the justifications related to Operational Continuity, frequent mainly in the bases for abstention; the application in other types of companies and economic sectors; the verification of the potential effects of the structure of the new audit report, established as of the base date 2016, in the manner in which the audit assertions are reflected in the justification for opinion changes. In addition, continuous reflection on the audit process and planning is encouraged in order to obtain further information that can be gathered to obtain increasingly complete and high-quality audit work.