Integrated Reporting Assurance : Perceptions of Auditors and Users in Spain

Integrated reporting is a key instrument used to inform stakeholders about the sustainability issues of a company. Only an assured report can effectively instill confidence in its users regarding the sustainability of the company. Based on the International Integrated Reporting Framework issued by the International Integrated Reporting Council (IIRC), the authors solicited perceptions from auditors and audit report users about several aspects of integrated reporting assurance. An analysis of the responses suggests that integrated reporting assurance is important, but there are many challenges (both methodological and related to the characteristics of non-financial information) for auditors to overcome. Reporting companies and auditors must work to overcome these problems. The former ones must improve the quality of non-financial information and the later must adapt their audit procedures. This paper provides valuable insights into preferences regarding the form and content of the audit report on integrated reporting. This study is useful to regulators of audit activity, auditors’ corporations, the IIRC, and other international associations, academics, and audit report users, and contributes to the current integrated reporting literature by examining the perceptions of auditors and users regarding the assurance of integrated reporting. Integrated reporting assurance is still an under-explored field of research.


Introduction
After the financial crisis of 2008/2009, two tendencies emerged in corporate information-critics of the older reporting systems, and a growing interest in non-financial information [1][2][3][4].Non-financial information in the form of sustainability reports, corporate social responsibility (CSR) reports, and other reports is used to complement classical financial accounting information (financial statements and management reports) by communicating a firm's principal environmental and social performance indicators as well as their related risks and opportunities.We use the denomination of CSR reports or sustainability reports to refer to a set of predominantly non-financial reports, except the integrated reports.These reports may also be referred to as environmental, social, and governance (ESG) reports, corporate responsibility reports, triple bottom line (social, environmental, and financial) reports, etc.Many institutions have established guidelines for CSR reporting, with the Global Reporting Initiative (GRI) being the most popular guide [5,6].There are many other international initiatives regarding guidelines for non-financial information.We also want to highlight The Carbon Disclosure Project (CDP), The European Federation of Financial Analyst Societies (EFFAS), The American Sustainability Accounting Standards Board (SASB), and The International Organization for Standardization (ISO).The proportion of leading companies reporting CSR continues to grow, with approximately three-quarters (75 percent) of the top 100 companies from 49 countries (N100) surveyed in 2017 producing such reports [6].However, criticisms about non-financial information have also been raised.There is a risk of "greenwashing" with CSR reports [7,8].Additionally, preparing CSR reports or sustainability reports as separate non-financial reports creates information dissemination, "more pages unread", and confusion for stakeholders [9].
To minimize these problems and to increase stakeholders' trust, the IIRC launched a principlebased framework for integrated reporting [10].This framework proposes an integration of financial and non-financial information [11].Integrated reporting is the latest development in sustainability reporting [12]; it should explain how an organization generates value over time using social, environmental, and economic factors, as well as the link between risk, strategy, and the entity's business model [10].Some authors argue, nevertheless, that integrated reporting has moved from a sustainability focus that took into account all stakeholders to a focus that prioritizes investors [13,14].
To enhance the credibility of non-financial information and integrated reporting, there is a need for assurance [5,[15][16][17].The decision usefulness of integrated reporting must be enhanced [18] if the practice is to be widely adopted in the future [11,19].Nevertheless, many challenges to the assurance of non-financial information have already been identified in the literature, principally the following: the absence of mature reporting systems for collecting and processing non-financial information [5,16], the subjectivity of some contents of integrated reporting [20], the lack of suitable professional standards to issue an opinion over some contents of integrated reporting [5,16,21], and other methodological challenges to assurance [19,20].This study provides confirmatory evidence that these difficulties must be solved before auditors can assure integrated reporting.However, from the perspective of stakeholders, it is essential for such information to have the same guarantees as traditional financial information.The survival of non-financial information depends on its credibility and usefulness [11,19,21].
Our study seeks to contribute to the integrated reporting assurance debate and responds to calls from the IIRC and International Auditing and Assurance Standards Board (IAASB) for more innovative assurance research to fulfil the actual reporting needs of organizations [22,23].Although research on integrated reporting is available [24], the number of empirical studies regarding assurance is still very small, with limited samples or limited questionnaires [18,20,25].
We solicited auditors' and audit report users' (henceforth "users") perceptions in Spain through an anonymous online survey in order to address the following research questions (RQ): • RQ1-What are auditors' and users' perceptions of the importance of assuring the content of integrated reporting?• RQ2-What are auditors' and users' perceptions of the form and content of an audit report on an integrated reporting?• RQ3-What are auditors' perceptions of the challenges of assuring integrated reporting?
We present analyses of the perceptions of 103 auditors and 109 users as a full sample and we also present both groups separately.Regarding the questions only answered by the auditors, we present analyses of their perceptions as a full sample and also separately along two dimensions (rank in the audit firm and years of professional experience).Overall, we found that auditors and users considered it important to audit the entire content of integrated reporting.The most important content elements to assure are "financial performance", "basis of preparation and presentation" and "governance".Collectively, they believed that two separated audit reports should be issued by auditors-one for annual accounts and another for the remaining information contained in the integrated reporting.With regard to the assurance level for non-financial information in the audit report, they chose reasonable assurance.Respondents from both groups noted the relevance of describing the inherent limitations of integrated reporting in the conclusions of the audit report.With regard to the questions only answered by auditors, they noted a high degree of difficulty in auditing the following content elements of integrated reporting-"outlook", "non-financial performance", and "strategy and resource allocation".When asked about the most important challenges of auditing integrated reports, auditors cited the lack of generally accepted standards for reporting non-financial information, as well as the inability to adopt a control reliance strategy regarding non-financial information generation systems.Finally, they also considered the current state of responsibility for assuring integrated reporting to be unacceptable.
Our study makes several important contributions.It provides evidence about actual perceptions of auditors and audit report users regarding the importance of assuring integrated reporting.It also describes the needs of both groups regarding the form and content of the audit report on integrated reporting.This information could be valuable for the IIRC, the IAASB, and other regulatory agencies anticipating the possible implementation of mandatory integrated reporting and the associated mandatory audits.Future modifications of the audit report should seek an agreement between auditors and users to guarantee that the changes are successful [26].In addition, our findings may be useful to the audit practice because they could be an important step toward focusing on-and thus solving-the difficulties that auditors face when assuring integrated reporting.Some of these difficulties are methodological and must be solved by the auditors themselves.Other difficulties pertain to some aspects of integrated reporting that are still not well defined in the International Integrated Reporting Framework and must be worked out by the IIRC.Other challenges stem from the apparent lack of rigor of non-financial information.Companies have to improve their controls over information flows to make non-financial information more reliable [5].Finally, we contribute to the literature on integrated reporting and assurance.Assurance on integrated reporting remains an under-explored research topic [18,20,25,27].Our study illustrates some areas that are the most likely to benefit from future research.
The remainder of the paper is structured as follows.Section 2 provides the background and basis for this study.Section 3 describes the methodology, including the participants and survey development.Section 4 presents and discusses the results.We conclude with a summary of our findings, the study's limitations, and suggestions for future research.

Background and Literature Review
In this section, the extant literature on integrated reporting is reviewed under two broad headings; first is integrated reporting, and second is the assurance on integrated reporting.

Integrated Reporting
The transformation of corporate reporting over time has been very important.Stakeholders decreased their confidence in traditional financial reporting [2,3] and demanded that organizations provide non-financial information in the form of sustainability reports or CSR reports [1,4].As noted earlier, the number of CSR reports has grown significantly over recent years, with 75% of the N100 companies and 93% of the G250 companies issuing CSR reports [6].KPMG's survey concludes that "corporate responsibility reporting is standard practice for large and mid-cap companies around the world" [6] (p.4).However, the information provided by companies is often disconnected and static [28], which complicates its analysis and reduces its value and decision usefulness [9].There is also a risk that these reports are not effective and instead serve as mere public relations exercises [7,8].Finally, there is a risk of information overload [29].
The IIRC ["a global coalition of regulators, investors, companies, standard setters, the accounting profession, and NGOs", while its long-term vision is "a world in which integrated thinking is embedded within mainstream business practice in the public and private sectors, facilitated by integrated reporting as the corporate reporting norm" [10] (p.1)], created in 2010, proposed that the solution to these problems is to integrate all the information issued by organizations in one brief report that, without forgetting the past, covers issues such as strategy, forward-looking information, and economic, environmental, social, and governance issues [10].Integrated reporting and integrated thinking, as well as a focus on value creation, should be the next stage of the evolution of corporate reporting [10, 12,30].
There is a growing interest in non-financial information, sustainability information, and integrated reporting [16].In March 2010, South Africa became the first country in the world to require listed companies to issue an integrated report with the effective application of the King III initiatives [31].Moreover, some stock exchange markets, such as Sao Paulo, Singapore, Kuala Lumpur, and Copenhagen, urged companies to report information about environmental, social, or governance issues [32].Some countries in the South Asian region, one of the fastest growing regions in the world economy, are moving towards integrated reporting [33].With the adoption of the Directive 2014/95/EU (the so-called "CSR Guideline") and EU Guidelines 2017/C215/01, large entities of public interest in the European Union (EU) have to publish a non-financial statement for the fiscal year beginning on or after 1 January 2017.The non-financial statement should contain aspects relating to (at minimum) environmental matters, social and employee-related matters, respect for human rights, anti-corruption, and bribery matters [34,35].Recent research shows a tendency for companies located within a member state of the EU to replace the independent sustainability report in favor of integrated reporting [29].Even public sector organizations have begun to adopt integrated reporting in order to improve their accountability [36].
The IIRC has developed an International Integrated Reporting Framework to support companies in preparing an integrated reporting.It is defined as "a concise communication about how an organization's strategy, governance, performance and prospects, in the context of its external environment, lead to the creation of value over the short, medium and long term" [10] (p.7).Integrating information requires more than simply combining the annual accounts with other reports.An integrated report should reflect the systemic thinking of the organization and tackle the performance of a company by taking into account all the capitals that affect value creation over time [10].The IIRC's Framework [10] identifies six types of capital-namely financial, manufactured, intellectual, human, social and relationship, and natural.Internationally, over 1500 companies have thus far adopted integrated reporting practice [37].The KPMG survey states that "the number of companies that specifically label their reports as integrated is growing slowly but steadily" [6] (p.24).The content and presentation of integrated reporting are underpinned by the following guiding principles: strategic focus and future orientation, connectivity of information, stakeholder relationships, materiality, conciseness, reliability and completeness, consistency, and comparability [10].Furthermore, IIRC's Framework [10] comprises eight content elements, thus incorporating the following: organizational overview and external environment, governance, business model, risks and opportunities, strategy and resource allocation, performance (financial and non-financial), outlook, and basis of presentation and presentation.
Beyond the value of issuing integrated reports, companies will benefit from developing integrated thinking [16,30,38].They will focus on financial and non-financial performance and especially on the interrelation between the two [28,38].Companies will develop a clear vision of the central factors that affect value creation and how they contribute to their strategic goals and to developing new opportunities [39].The communication of these goals to employees will enhance their collaboration within and their commitment to the organization, as they will gain a better understanding of the company for which they are employed and how it generates and maintains value over time [40].Integrated information should be a powerful tool to show that a company is committed to sustainability and to having a positive impact on society; thus, integrated reporting should enhance the company's reputation [1,[41][42][43] and generate more confidence among stockholders [44].Integrated reporting potentially represents the most significant change to corporate reporting seen in years [45].Integrated information should contribute to more effective decisions by providers of financial capital, thus enhancing the performance of their investments and achieving a more efficient allocation of their resources [9,19] and of social, natural, and human capitals [46].Moreover, it should contribute to economic stability and a decrease in capital costs for reporting companies, as it would result in transparency for all market agents [40], better decision-making processes, long-term, holistic, and well-balanced views of all stakeholders, and effective risk identification [10,11].Nevertheless, some critical voices have also been raised against the integrated report as an effective sustainability disclosure tool, as it is principally investor focused [13,14].

Assurance on Integrated Reporting
There is a growing trend in assuring non-financial information; 67% of the G250 companies that issued CSR reports in 2017 let them be verified by third parties (in 2008, only 40% of them did so) [6].The credibility of non-financial reports-and therefore of the informing organizations-is enhanced by assurance [16,19,38,41,43,47].For example, 91% of the reports published by companies located in a member state of the EU and provided in the "Getting Started" section of the Integrated Reporting Example Database were assured by independent third parties [29].The IIRC has established-in cooperation with Black Sun Plc, a stakeholder communications company-a platform on which English language integrated reports are available, and the database has three central sections, the "Getting Started" section, the "Recognized Reports" section, and the "Integrated Reporting Reporters" section.Stakeholders have demanded the same guarantees on integrated reporting as they have on traditional financial information [22,48].If such guarantees are not provided, the survival of integrated reporting will be at stake.Credibility is one of the keys to ensuring the widespread acceptance of integrated reporting by stakeholders [11,19].Integrated reporting requires integrated assurance [16,19].In the interest of brevity, we do not distinguish in our study between "audit" and other "assurance" services.In addition, we do not examine differences between audits of financial statements per International Standards on Auditing (ISA), limited and reasonable assurance engagements per International Standards on Assurance Engagement (ISAE), and reviews of information per International Standards on Review Engagements (ISRE).On the other hand, we do not use the term "integrated assurance" as it is used in Auditing Standard no.2201: "An audit of internal control over financial reporting that is integrated with an audit of financial statements" [49], but rather in the sense used by Eccles et al. [19].However, the road to achieving this assurance is still long because there are many difficulties to be addressed before auditors can issue an audit report on the entire content of an integrated report.Challenges faced by auditors when assuring non-financial information are multiple, and they derive mainly from specific attributes of non-financial information.Principally, the absence of robust non-financial information generation systems and the lack of generally accepted global standards for measuring and reporting non-financial information [5,16].However, there are also some methodological challenges that audit activity has to overcome [5,16,[19][20][21]50].
We do not expound on the question about who should provide integrated assurance.Like other authors, we consider the Big 4 (Deloitte, EY, KPMG, and PWC) and others such as BDO International and Grant Thornton International to have the resources, professional skills, multidisciplinary teams, and reputations to perform the necessary audits [5,19].A total of 85% of the reports published by a company located in a member state of the EU and provided in the "Getting Started" section of the Integrated Reporting Example Database were assured by one of the Big 4 [29].
One of the main issues to address is the quality of non-financial information.The source of this information is not typically robust information systems or processes, and thus, the information is not subject to the same controls as financial information [5,16,19].Those systems or processes normally do not assure the integrity of the information.Auditors are used to apply a control based audit approach.Will they be able to use such methodologies and procedures on non-financial information?There are some doubts as to whether it is possible to assure the content of integrated reporting without assuring the information generation process [11,50].We are not saying that non-financial information is inherently inaccurate, but companies will have to improve their internal controls over non-financial information flows and professionalize their information systems so that auditors can test them and rely on their controls [5].Audit committees and other systems of corporate governance will play a key role [50,51], but they are not a focal point of this study.
A complicating factor is the absence of generally accepted global standards for measuring and reporting non-financial information, i.e., standards similar to International Financial Reporting Standards (IFRS) or United States Generally Accepted Accounting Principles (US GAAP) for financial information.Only when the content and presentation of non-financial information is standardized will information be sufficiently comparable and meaningful for stakeholders and will auditors be able to issue their opinions on it.With this goal in mind, local standards have to be reviewed and homogenized, and new international standards have to be issued [1,11,19].The EU "CSR Guideline" is a key step toward increasing the importance of CSR [6] and, therefore, of integrated reporting, but it may be too flexible.Moving towards an international framework would both streamline the reporting process for new entities and increase consistency between reports.
Assurance engagements over non-financial information are normally performed in accordance with ISAE (issued by the IAASB) or AA1000AS (issued by AccountAbility) [20].Other standards are also applied, such as ISO 19011 and ISO 14064-3 (issued by the International Standards Organization) and AT201 and AT701 (issued by the American Institute of Certified Public Accountants), but these are used less than ISAE and AA1000AS in Spain.The IIRC [20] can be consulted for a detailed review of all standards that could be used to assure non-financial information.ISAE and AA1000AS are designed to provide assurance on information that can be objectively tested according to clearly defined criteria [52,53].Therefore, they are not completely appropriate to issue an opinion over some of the contents included in an integrated report, as some of them can only be described and not monetized or even quantified [5,16,20,22].Maroun [20] proposes an interpretative assurance model that "focuses on the methods and processes used to support management representations instead of using predefined procedures (as per ISA and ISAE) to test accuracy of the data itself" [20] (p.418).
Another set of challenges that auditors will have to face in order to issue an integrated assurance opinion are methodological [5,15,19,22,54].Which level of assurance are auditors able to provide on the entire content of an integrated report?How can materiality on non-financial issues be fixed?How can the scope of integrated reporting, the connectivity, and the integrity of information and the narrative part of the report (including forward-looking information) be audited?What should be the content and form of an integrated audit report?How should the report inform about the work of other experts?We must ask ourselves whether these challenges and the threats of litigation and reputation loss will restrain the evolution of audit activity in this field [19].
There is still debate over the level of assurance auditors should (or could) issue for all of the information contained in an integrated report [16,22,50].Perhaps the framework of "reasonable" and "limited" assurance is not suited for integrated reporting [16,20].Nevertheless, it represents a first step toward making the assurance of integrated reporting possible.Of the reports published by a company located in a member state of the EU and provided in the "Getting Started" section of the Integrated Report Example Database, 57% were assured with reasonable assurance and 33% with limited assurance; 10% of the integrated reports were based on varying assurance intensity.While an entire integrated report is generally audited with a reasonable assurance level, individual parts are based on limited assurance [29].This disparity might impair the credibility of integrated reports.In our opinion, reasonable assurance on the entirety of the information in the integrated report would be ideal, as it would increase the value of integrated report for the stakeholders during decision-making [19].The audit opinion would be easy to understand, as the assurance level would be the same as what we already have for financial information [22].
Materiality is a key concept in financial and non-financial information reporting, as well as in the audit processes [19,55].ISA 320 "Materiality in planning and performing an audit" guides auditors in determining materiality but only for financial statements audits [55].No methodology is defined in the IIRC's Framework to fix the risk of misstatement in integrated reporting, as no common unit of measurement can be found for reporting issues [5,20,54].The challenge of auditors will be to combine the materiality of financial reporting frameworks and integrated reporting framework in order to define a single materiality for integrated reporting assurance [22,54].As we outline below, auditors will need the help of many experts for many different reasons.The importance of considering qualitative factors will increase [19,20].The need to broaden the knowledge of the client (especially of the value creation model) will be crucial.Therefore, a much broader set of skills will be required [16].
The connectivity of all the information reported by organizations is one of the main purposes of integrated reporting [10].Determining whether such connectivity is "sufficient" or "correct" is one of the main challenges auditors will face [11].The links between the integrated reporting and other reports will also be a source of difficulty for auditors.Do those links cross the boundaries of integrated reporting?Will this information also be assured when the integrated report is audited [16]?Assuring the integrity of the information will also be difficult [11].Striking a balance between integrity and conciseness will be complex for organizations.It will also be challenging for auditors to ensure that the selected information is complete.
Criticisms of integrated reporting are mainly related to its use of narrative and forward-looking information [11,56,57].The credibility of the disclosed qualitative and quantitative forward-looking information could be improved through the effective assurance of integrated reports.To audit such information, auditors will have to exercise a great degree of professional judgment and scepticism [22].Determining what is sufficient evidence regarding non-financial information is still a topic of debate [16,19,21].Auditors will have to judge whether reports' contexts and tones are appropriate so that readers are not misled [19].Current assurance frameworks do not provide guidance regarding these issues.Could auditors assume those risks?
Integrated reporting includes information about many aspects of the organizations, their environment, and their value creation model [10].We are of the opinion that an integrated vision of the audit should be adopted.Companies would be better advised by a unique multidisciplinary audit team rather than multiple teams from one or various audit firms [10, 16,58].Nevertheless, it will be very difficult and expensive to bring together IT experts, financial report experts, experts in the sector of the client, experts on sustainability, and environmental experts.Although some authors argue that a competitive assurance service market would offer assurance services in a cost-efficient way [5], the cost of involving experts could make the cost of auditing too high [16,22].Are we willing to pay these experts?Auditors will have to rely on the work of other experts in order to express an opinion on all the disclosures of the integrated reporting.Therefore, international auditing standards regarding this matter will have to be adapted to the special characteristics of integrated reporting.We are thinking of ISA 600 "Special considerations-audits of group financial statements (including the work of component auditors)", ISA 610 "Using the work of internal auditors", and ISA 620 "Using the work of an auditor's expert", issued by IASB.
Following Eccles et al. [19], we agree that an integrated audit opinion should be issued on the complete content of an integrated reporting.A specific standard should be issued regarding the content and form of an integrated audit opinion, and the following questions should be addressed.How should the opinion describe the procedures used?How should the opinion communicate the inherent limitations of integrated reporting?How should the opinion consider the work of other experts?How should organizations' and auditors' responsibilities be described?What should be the content and form of audit conclusions [11,19,50]?Should the audit report be in a long or short form [22]?There are still no examples of an assurance opinion that covers the complete content of an integrated report [20].With respect to the conclusions in the audit opinion, auditors assure in their actual audit report that financial statements give, in all material aspects, the true and fair view of the economic and financial position of the company (the concept "true and fair view", as defined by ISA 700 "Forming an opinion and reporting on financial statements" [59]-is a concept that is more common outside of the United States, where the term "presents fairly" is used).Stakeholders will demand that auditors assure that integrated reporting gives "a true and fair view of an organization's sustainability and its capacity to create value over time" [11,19].Two challenges will have to be overcome-how to determine the "true and fair view of sustainability", and how to audit the long-term capacity of an organization.Academic literature shows that the standardization of audit opinions is a problem for users [26,[60][61][62].Nevertheless, we are of the opinion that a certain level of standardization is needed to assure comparability of audit reports at an international level.Free format audit reports, as considered in ISAE 3000, are generally complicated to analyze and to compare.
To answer some of the questions outlined above, we conducted a survey of auditors and audit reports users.We asked about their perceptions of the importance of auditing integrated reporting and about the form and content of a possible subsequent audit report.Focusing on auditors, we asked them about the challenges they expect to face when assuring integrated reports.The next section describes our methodology.

Methodology
To address our research questions and to gain insight into integrated reporting assurance, we solicited auditors' and users' perceptions through an anonymous online survey.We purposefully selected, based on availability and representativeness, 118 auditors and 126 users in Spain.This selection ensured that the chosen auditors had at least three years of experience and users had five years of experience in their respective fields; both groups also had a detailed understanding of the relevant reporting and assurance frameworks and a solid appreciation of the format and content of an integrated report [63].Although the sample was not a random sample, it was sufficiently representative to gain insight into auditors' and users' perceptions in Spain regarding the assurance of integrated reporting.In total, 103 auditors and 109 users completed the survey, which represented response rates of 88% and 87%, respectively.Although it was arduous to obtain these high response rates, they ensured that the sample was representative.We took care that the selection of the sample did not introduce any bias into our investigation.Table 1 provides demographics for our sample of 212 participants.In total, 64% of our participants were male and 36% were female.Approximately 30% of our respondents had less than 12 years of professional experience, and 70% had 12 years or more of experience.Regarding education level, 65% had a bachelor's degree.The majority of auditors (89%) belonged to the Big 4.In terms of rank in the audit firm, 21% were seniors, 44% were managers, and 35% were partners.Users, on the other hand, were mainly CFOs (46%), 15% were CEOs, and 39% worked in other fields.This led to a very adequate sample for the objectives of our investigation.The sample was prepared ad-hoc for our study.
We used the Chi-square test to measure the dependence between being an auditor or user and the following demographics: gender (p-value > 0.05), professional experience (p-value < 0.05), and education level (p-value < 0.05).The higher percentage of auditors with fewer than 12 years of professional experience in comparison with users is explained by the pyramidal structure of audit firms.On the other hand, the Big 4 in Spain offer junior-level auditors master's degrees during the first years of professional experience.This fact could explain the higher percentage of auditors with a master's degree in comparison with users.Anyway, we took into account the number of data points in order to ensure the tests were applicable in the whole study.
An earlier version of the survey instrument was reviewed by five practicing auditors, all at high professional levels (senior, manager, and partner), and by five users who were experts on financial topics.They were requested to evaluate the questionnaire based on diverse aspects such as clarity or ambiguity in the exposition of the issues as well as its relevance to the field of study.They were also asked to make comments regarding the time for completing the questionnaire, the sequence of the questions, and the capacity to contextualize the instructions.Pre-test of the survey is often used to solve problems before the survey goes into the field [64,65].Based on the feedback received, the final instrument was improved and administered online to survey participants.The 212 valid questionnaires were completed from April to October 2016.We do not present the whole survey because of its length.A hardcopy of the online survey is available upon request.
The questionnaire began with individual data on the respondents, reported in Table 1.In all, it contained five questions (questions 1 to 5) for all participants and two additional questions (questions 6 and 7) only for auditors.Questions are detailed in Table 2.

Question Number
Questions Possible Answers 1 Assess the importance of auditing the complete content of integrated reporting if the integrated report is adopted as the corporate reporting model.
Five-point Likert scale with 1 equaling "not important" and 5 indicating "very important" 2 Assess the importance of auditing the content elements proposed by the IIRC for the integrated report if the integrated report is adopted as the corporate reporting model.
Five-point Likert scale with 1 equaling "not important" and 5 indicating "very important" 3 If the integrated report is adopted as the corporate reporting model and the integrated report is audited, which type of audit report should auditors issue?
Two separated audit reports One single integrated report 4 If the integrated report is adopted as the corporate reporting model and the integrated report is audited, which assurance level would you consider appropriate for following aspects: non-financial information is free of material misstatements due to errors; non-financial information is free of material misstatements due to fraud.

No assurance needed Limited assurance Reasonable assurance
Total assurance 5 Asses the relevance of including the following information in the audit report on the integrated report: description of work done to audit non-financial information; description of inherent limitations of integrated reporting of the organization; conclusion that the integrated reporting provides a true and fair view of the organization's sustainability; conclusion about the capacity of the organization to create value over time.
Five-point Likert scale with 1 equalling "not relevant" to 5 "very relevant" 6 Asses the difficulty of auditing the content elements proposed by the IIRC for the integrated reporting.
Five-point Likert scale with 1 equalling "no difficulty" to 5 "much difficulty" 7 Assess following challenges regarding if they complicate auditing integrated reports: inability to adopt control reliance strategy regarding non-financial information generation systems; lack of generally accepted standards to report and measure non-financial information; inability to meet expectations of stakeholders regarding reasonable assurance on non-financial information; lack of generally accepted definition of materiality for integrated reporting; determination of integrated reporting boundary; assessment of connectivity and completeness of information in integrated reporting; high degree of professional judgment and scepticism; costs exceed benefits; responsibility is unacceptable.
Five-point Likert scale with 1 equalling "does not complicate auditing integrated report to 5 "complicates very much auditing integrated report" Questions 1 and 2 used a five-point Likert scale with 1 equaling "not important" and 5 indicating "very important".The questions were related to the perceptions of the participants on the importance of auditing the integrated report and its concrete content elements.Questions 3, 4, and 5 were related to the perceptions of participants on the form (integrated audit report or separated audit reports) and content (assurance level and conclusions) of the audit report on the integrated report.Questions 6 and 7 were posed only to auditors; they used a five-point Likert scale and were related to auditors' perceptions of the difficulties of auditing the content elements of integrated reporting.Auditors were also asked to assess nine challenges identified by the literature in order to explore the origins of auditors' difficulties in assuring integrated reporting.

Discussion of Results
In this section, we present and discuss survey results for RQ1, RQ2, and RQ3.Unless we indicate otherwise, all survey questions were rated on an ordinal scale of one to five.Mean tests are not appropriate when using ordinal data [66] or when data are skewed [67].In all cases, means and medians were directionally consistent.We performed the parametric tests corresponding to the non-parametric tests shown and obtained very similar results.

RQ1: The Importance of Assuring Integrated Reporting
Table 3 presents our findings for the perceptions of our respondents regarding the importance of assuring the complete content of integrated reporting and its concrete content elements.We assessed mean and median responses for our full sample, and we also considered differences between the groups (users and auditors).Respondents considered it important for the entire content of integrated reporting to be audited.The median and mean were above the central value (3) for the whole sample and in each group.These results are encouraging because prior literature emphasises the importance of assuring integrated reporting [5,11,15,16,19], as we outline above.
Regarding the specific content elements of the integrated reporting, the full sample, as well as its constituent groups, considered auditing "financial performance", "basis of preparation and presentation", and "governance" to be the most important content elements to assure.On the other hand, auditing "organizational overview and external environment", "non-financial performance", and "strategy and resource allocation" were considered least important by the full sample as well as by both groups.
Comparing users to auditors, we found one notable difference.Auditors attached higher importance to auditing the whole integrated report as well as all specific content elements (auditors' means exceeded users' means in all cases).All differences regarding the importance of assurance were significant.Auditors attached higher importance to auditing the "basis of preparation and presentation" (mean = 4.18).This part of the integrated report described the organization's materiality and reporting boundary determination process [10].As these two items determine the content of the integrated report and therefore what auditors have to audit, their interest in assuring this part of the integrated report is reasonable.Surprisingly, respondents-especially users-attributed relatively low importance to auditing "non-financial performance".Means and medians were around the central value for the whole sample (mean = 3.14) and for each group (mean = 3.01 by users and mean = 3.28 by auditors).Prior literature suggested the importance of stakeholders having the same guarantees on non-financial information as on financial information [11,19,22].We did not explore the reasons for these values.Regarding auditors, we can attribute the low value to the difficulty of auditing non-financial information and the fact that auditors want to issue different assurance levels for financial and non-financial information, as discussed in Table 5.There are different possible explanations for users' negative perception regarding the importance of assuring "non-financial performance".Non-financial disclosures are not always relevant for some important stakeholders [68,69].The link between many important non-financial metrics and organizations' key risks and strategic objectives is often difficult to understand [70].Or, perhaps users are aware of the difficulties of assuring the content of an integrated report, and they know that any opinion about it would just be a commentary or personal view [20].To resolve this question, it would be necessary to explore, in depth, the respondents' motivations for being indifferent to the assurance of "non-financial performance" in the integrated report.Overall, our results suggest that auditors are more interested than users in the auditing of integrated reporting.Next, we examine respondents' perceptions of form and content of audit reports on integrated reporting under RQ2.Notes: a We used the Mann-Whitney U test to compare users (N = 109) versus auditors (N = 103).All p-values were two-tailed and we considered p-values equal to or less than 0.05 as significant; b Participants rated these items on a scale of 1 (not important) to 5 (very important).

RQ2: Form and Content of Audit Report on Integrated Reporting
Table 4 summarizes the findings for respondents' perceptions regarding the form of the audit report on integrated reporting.We assessed the frequencies of responses for the full sample and for the two groups (users and auditors).Notes: a All p-values were two-tailed and we considered p-values equal to or less than 0.05 as significant.b The question in the survey was: "If the integrated report is adopted as the corporate reporting model and the integrated report is audited, which type of audit report should auditors issue?".
Overall, respondents preferred two separate audit reports, one for the annual accounts and the other for the rest of the information reported in the integrated reporting.Prior literature suggested that a single integrated audit assurance opinion would be more desirable [19].Comparing auditors to users, we found a major difference (the Chi-square test revealed a significant difference).Auditors had a clear preference for issuing two separated audit reports (frequency = 66%).The reason could be that they want to assure financial and non-financial information with different assurance levels, as we discuss in Table 5.This would not be possible if a single integrated audit report were issued.Another reason may be that they are aware of the difficulty of handling a long opinion with such diverse content.Users, on the other hand, did not have a clear preference for either of the two options (frequencies were approximately 50%).In future studies, it would be interesting to examine if the reason for the users' lack of preference lies in the fact that they are not used to non-financial information assurance, in the lack of information about the consequences of having two separate audit reports to analyze, or in other reasons.
Table 5 summarizes our findings for respondents' perceptions regarding the appropriate assurance level in the audit report for non-financial information.We assessed the frequencies of responses for the full sample and for the two groups (users and auditors).Notes: a All p-values were two-tailed and we considered p-values equal to or less than 0.05 as significant.b The question in the survey was: "If the integrated report is adopted as the corporate reporting model and the integrated report is audited, which assurance level would you consider appropriate for following aspects?".
Regarding the assurance level in the audit report for non-financial information, the majority of our respondents chose reasonable assurance (51.9% regarding errors in non-financial information and 44.3% when asked about fraud).International audit frameworks such as ISA dictate that financial information must be audited with a reasonable level of assurance.If we follow our survey results, non-financial information should be assured with the same level of assurance as that currently used for financial information.As we noted before, this would be, in our opinion, the best solution for the audit report on integrated report.Nevertheless, a relatively high percentage of our participants (30.2%) would prefer limited assurance in the audit report with regard to non-financial information.Prior literature has shown many non-answered questions regarding the possibility of combining different assurance levels in a unique report [5,18].
We found significant differences between auditors and users.If we concentrated on errors in non-financial information, the percentage of auditors who would prefer limited assurance was higher than the percentage of users with that preference (37.9% versus 22.9%).This difference was accentuated when participants were asked about fraud regarding non-financial information.The majority of auditors (43.7%) preferred limited assurance versus 17.4% of the users.As we analyze in Table 8, the characteristics of non-financial information hindered auditor respondents in their assurance of integrated reports.Not surprisingly, a high percentage of them only wanted to issue an opinion with limited assurance.To more deeply investigate the reasons for the differences in the results when comparing errors to fraud, it would be interesting to analyze the incentives for fraud associated with integrated report assurance engagements and how they impact the risk of material misstatement for integrated reports [5].The differences in responses between auditors and users regarding assurance level in the audit opinion also show that the expectation gap still exists.Nevertheless, that gap is not the focus of this study.
Table 6 presents findings for items concerning the relevance of including some content in the audit report on integrated reporting.The content of the audit report on financial statements is internationally standardized [59].We presented the survey's participants with four additional content elements to include in the audit report.We assessed the mean and median responses for our full sample, and we considered differences by groups (users and auditors).Respondents considered it relevant to include the four proposed information items in the audit report on an integrated reporting (median and mean were above the central value (3) for the whole sample and in each of the groups).Comparing users to auditors, we found one notable difference.With regard to the specific items, users considered the conclusion regarding the true and fair view of the sustainability of the company to be the most relevant (mean = 4.46).This finding is consistent with prior literature [11,19].Auditors, not surprisingly, were more interested in emphasizing the inherent limitations of integrated reporting (mean = 4.40); they want to emphasize the challenges they face when assuring an integrated report.Notwithstanding the differences between the two groups, our respondents considered all proposed items to be relevant.In our opinion, the four proposed content elements should be included in the audit report.Next, we examine auditors' perceptions of the challenges of integrated report assurance.Notes: a We used the Mann-Whitney U test to compare users (N = 109) versus auditors (N = 103).All p-values were two-tailed and we considered p-values equal to or less than 0.05 as significant; b Participants rated these items on a scale of 1 (not relevant) to 5 (very relevant).

RQ3: Challenges Of Assuring Integrated Reporting
Table 7 summarizes findings for auditors' perceptions regarding the difficulty of auditing the content elements of the integrated report.We assessed mean responses for all auditors, and we considered responses by rank (partners, managers, and seniors) and by years of professional experience (fewer than 12 years and 12 years or more).
Respondents reported a relatively high level of difficulty auditing all the content elements of integrated reporting.All means were above the central value (3) for the full sample of auditors except for the "basis of preparation and presentation" (mean = 2.92).Not surprisingly, auditors saw the most difficulty in auditing "outlook" and "non-financial performance" (mean = 4.42 and 4.17, respectively).We found no major differences when comparing auditors' perceptions by rank or by years of professional experience.Nevertheless, auditors with 12 years or more of professional experience saw a higher difficulty in assuring the content elements of integrated reporting than those with less experience did, with the exceptions of "governance" and "basis of preparation and presentation".One explanation is that experience and professional training make highly experienced auditors more aware of the difficulties of assuring integrated reporting.We next consider the possible motives for these difficulties.
Table 8 presents findings regarding auditors' challenges when auditing integrated reporting.We assessed mean responses for all auditors, and we considered responses by rank (partners, managers, and seniors) and by years of professional experience (fewer than 12 years and 12 years or more).Auditors reported a relatively high level of difficulty in the presented challenges (all means were above the central value (3) for the full sample of auditors, as well as in the groups by rank and by professional experience).Auditors noted that the most difficulties arise from two characteristics of non-financial information already identified in the literature-the lack of generally accepted standards for reporting and measuring non-financial information (mean = 4.38), and the inability to adopt a control reliance strategy regarding non-financial information generation systems (mean = 4.22) [5,11,16,19].Following our survey's results, we encourage companies to improve their non-financial reporting systems so that auditors can rely on them.In addition, auditors stated that their high level of responsibility also hinders them to a great degree when assuring an integrated report (mean = 4.24).This finding is consistent with the concerns raised by Eccles et al. [19] regarding the increase in litigation risk (and reputation loss) for accounting companies if prominence is given to non-financial information.The authors argue that regulation should create "an environment that encourages the accounting firms to make the investments necessary to provide high-quality assurance on non-financial information" [19] (p.10).Notes: a We used the Kruskal-Wallis test to compare partners (N = 36) versus managers (N = 45) and seniors (N = 22).All p-values were two-tailed and we considered p-values equal to or less than 0.05 as significant; b We used the Mann-Whitney U test to compare auditors with fewer than 12 years of professional experience (N = 42) versus auditors with 12 or more years of professional experience (N = 61).All p-values were two-tailed and we considered p-values equal to or less than 0.05 as significant; c Participants rated these items on a scale of 1 (does not complicate auditing integrated report ) to 5 (complicates very much auditing integrated report); d All partners (N = 36) had 12 years or more of professional experience.Twenty-one managers had fewer than 12 years of professional experience, and twenty-four had 12 years or more of professional experience.All seniors (N = 22) except one had fewer than 12 years of professional experience.
We found no major differences when we compared auditors' perceptions by rank.Nevertheless, partners generally reported being more hindered than managers and seniors by the proposed challenges (partners' means were, with two exceptions, above managers' and seniors' means).As noted before, audit experience may make partners more aware of the challenges of integrated reporting assurance.This assumption was confirmed when we analyzed the responses by years of professional experience.Not surprisingly, significant differences (except for one item) were found between the two groups.Auditors with more experience expressed a higher degree of concern regarding all proposed items than auditors with less experience did (means of the experienced auditors were above the other means for all items).The findings suggest the importance to auditors of formation in new reporting models as well as in the consequences for audit activity.The results also support the call for more research regarding non-financial information assurance challenges for auditors [5,20].If an integrated report's survival depends on its credibility, IIRC, auditors, researchers, and reporting companies must work to solve the assurance challenges.

Conclusions, Limitations, and Suggestions for Future Research
In this section, we conclude, discuss some limitations of this study, and outline some suggestions for future research.Our study provides valuable insights on the importance of integrated reporting assurance, audit report preferences, and auditors' challenges when assuring non-financial information.Participants' responses generally support the notion that integrated report assurance is important for users and auditors [5,11,15,16,19].Therefore, steps should be taken to address the assurance challenges already identified in the literature [5,16,[19][20][21].Our study provides confirmatory evidence that attributes of non-financial information as well as litigation risk are of particular concern to auditors.Therefore, reporting companies should work on their reporting systems to make non-financial information more reliable.Auditors, at the same time, have to adjust their methodologies, teams, and procedures to ensure audit activity develops parallel to corporate reporting.Our study also identifies differences and similarities in users' and auditors' perceptions of the importance of integrated reporting assurance and audit report preferences.The results also show differences and similarities (by rank and professional experience) in auditors' perceptions of challenges when assuring integrated reporting.
We did not analyze the presence of non-response bias [71].Nevertheless, our response rates were very high (near 90%).Our participants had different degrees of familiarity with audit activity, which could have influenced the quality of their responses.We tried to mitigate this problem with the pre-test of the survey instrument when problems with the understanding of some questions were detected.In addition, we did not consider this a real problem, as our intention was to analyze our respondents' perceptions and not the validity or authority of those perceptions.Users' answers could be made without adequate cost considerations [72].We tried to mitigate this problem by finding a common ground with the perceptions of auditors who were well aware of cost factors.
The study was conducted in a Spanish setting.While limiting the study to a single jurisdiction is an inherent limitation, in 2010, Spain was the country with the largest percentage of companies obtaining assurance opinions for their sustainability reports [19].Spain was also one of four countries with a significant increase (9%) in integrated reporting from 2015 to 2017 [6].Spanish auditors and audit report users are, therefore, in a strong position to provide detailed insights on integrated report assurance.
Despite the study's limitations, the results should be relevant for both academics and practitioners interested in providing assurance of integrated reports.The results should also be relevant to the IIRC and to other international associations.Non-financial information poses important challenges to assurance by auditors.Audit report users and auditors must work to overcome this problem.The former must improve the quality of non-financial information, and the latter must adapt their audit procedures.Only assured integrated reporting will have a future.
We also identify several areas that create cause for concern and would most likely benefit from future research.The three areas are related to the incentive structure of the surveyed groups.First, respondents reported not being especially interested in the fact that non-financial performance is assured.This issue particularly plagues users.As previously discussed, stakeholders need financial and non-financial information to make their decisions.It would be desirable for all information reported by a company to have the same guarantees and credibility.Second, users did not have a clear preference for any of the proposed audit report formats.Future studies should analyze the reasons for this lack of preference.Finally, it would be interesting to analyze the incentives for fraud associated with integrated report assurance engagements and how those incentives impact the risk of material misstatements for integrated reports.Respondents, particularly auditors, showed different assurance level preferences when asked about errors and fraud in non-financial information.
We also identify three issues that could be broadened in future research.First, a very interesting topic to study in future research and where there is a wide literature is the value relevance of non-financial information [73,74].This relevance will be diminished if we are not able to verify the non-financial disclosures.Second, it would be very interesting for future research to extend our study to non-professional investors.There might be significant differences when comparing their perceptions regarding the issues asked in the survey on the assurance of non-financial information with the perceptions of the surveyed users.Finally, it could be also very enriching for future research to separate respondents into three groups-auditors, users, and corporate reporting preparers.

Table 3 .
Perceptions of the importance of auditing the integrated reporting.

Table 4 .
Perceptions of audit report form.

Table 5 .
Perceptions of assurance levels in the audit report.

Table 6 .
Perceptions of the relevance of including some content in the audit report on integrated reporting.

Table 7 .
Perceptions of the difficulty of auditing the content elements of integrated reporting.We used the Kruskal-Wallis test to compare partners (N = 36) versus managers (N = 45) and seniors (N = 22).All p-values were two-tailed, and we considered p-values equal to or less than 0.05 as significant.b We used the Mann-Whitney U test to compare auditors with fewer than 12 years of professional experience (N = 42) versus auditors with 12 or more years of professional experience (N = 61).All p-values were two-tailed, and we considered p-values equal to or less than 0.05 as significant; c Participants rated these items on a scale of 1 (no difficulty) to 5 (much difficulty); d All partners (N = 36) had 12 years or more of professional experience.Twenty-one managers had fewer than 12 years of professional experience, and twenty-four had 12 years or more of professional experience.All seniors (N = 22) except one had fewer than 12 years of professional experience.

Table 8 .
Perceptions of auditors' challenges when auditing integrated reporting.