The aim of this project has been to contribute to the emerging field of "the sociology of accounting reliability" by examining the role of the auditor in the production of financial reports. This is a field which called for more research on how accounting reliability is created and, more fundamentally, what is understood by the term in different situations. In the project, we have approached this aim in two ways. First, we have conducted a longitudinal case study of a large listed company. Through this study we have been able to create knowledge of accounting reliability in an empirically grounded way, which has opened up new perspectives on the concept of accounting reliability and the role of auditing in this process. Second, we have supplemented this case study with a larger interview study of auditors in listed companies. Interviews have made it possible for us to visualize patterns in auditors' statements about their view of work with accounting reliability, which is not possible to do with a case study.

The three most important results from the project

The project has generated research results that, in addition to directly addressing the aim, also contribute to the research literatures on materiality, revenue recognition, market values and negotiations between auditors and companies that prepare the accounts. However, the most important results of the project are closely related to the aim of the study:

1.    The project can show that reliability, in accounting practice, is a multi-faceted concept that must be understood in its situation. Previous research has largely studied whether or to what extent the reporting is reliable or not, or, alternatively, how reliability should be defined and what the pros and cons of different definitions entail. What this study shows is that reliability, insofar as it is relevant to speak of it in singular, is a compound term that means different things in different situations and, therefore, when questioned, is motivated and justified in different ways depending on how the concept is questioned. In our case study, we find three main ways in which accounting reliability emerges: As a focus on (i) systems and initial production of numbers; (ii) the company's assessment of the business reliability of the numbers - where expenses and income must be properly recorded but costs and revenues not necessarily and automatically becoming correct, as in many cases this requires business assessments to be made; or (iii) how the numbers are communicated to investors and an interested public.
2.    Therefore, the role of the audit in the construction of accounting reliability cannot be understood as a questioning of a one-dimensional reliability, but as an challenging of one or more aspects of reliability. Auditors have limited resources available for their audit and the priorities they make can generate significantly different contributions to the company's production of reliability. When auditors make priorities that according to their methodology are based on assessments of materiality and risk, they simultaneously choose between focusing on systems and initial production of numbers, on the company's assessments of the business reliability of the numbers and how the numbers are communicated to investors and an interested public. The questioning of the client's choices and assessments often takes the form that the auditor "pushes" the client in a direction that is in line with the auditor's perception of compliance by asking questions. Auditors thus fill more of a heuristic function where the questioning is more about requiring justifications of the accounting figures rather than questioning the numbers as such.
3.    Another crucial factor contributing to how the auditor affects the construction of accounting reliability is how the auditor views his or her own professional role, as this is reflected in auditors' negotiation strategies and, as an implication of this, views on what is material. Many areas in a company's accounts, such as depreciation, provisions and measurements at fair values are subject to judgments and thus open to negotiations. Accountants can take advantage of this opportunity in different ways. They can engage in regular negotiations where different starting positions are taken, and a compromise is reached somewhere in between. Another option is that auditors reject this form of negotiation but can accept an "error" as long as the offset occurs against another "error" in the income statement and balance sheet. Another option is that they accept both ways of negotiating, or do not accept negotiation at all. This last option does not mean that auditors always impose their view on the client, but they will do so until convinced of the opposite. In this latter case, assessments may still be assessments, but what the auditor ultimately has to report is the opinion the auditor is most convinced of and not any negotiated compromise in-between. What option or combination of negotiation strategies that the auditor accepts as valid has direct effects on how to understand reliability in these situations but also indirect effects because these choices between bargaining strategies affect the space for and the role of materiality in an audit context.

New research questions

One of the main findings of the study is, as stated above, that accounting reliability is a multifaceted concept whose meaning depends on the situation in question. Since this research project has been based mainly on a case study of a company with many big projects, a new research question that must be asked as a consequence of this project is whether other situations, in other types of companies, will lead to additional implications of reliability. Previous research has shown that, in assessing goodwill, a view of accounting reliability that excludes management's insights—who ought to know best—and instead relies on independent actors’ opinions on what should be considered as the correct value of goodwill, is favoured. This project's study of major projects presents an opposite view, where management's insights are central to creating an account that is considered reliable. Future research focusing on additional issues in other types of companies could complement this image and make it possible to say more about in what contexts and situations that certain actors’ insights and opinions become more important than others.

Another issue that this project has brought about is the question of how auditors’ view of their professional role and which negotiation configurations that are considered acceptable are reflected in the design and implementation of an audit. This is not an easy question to study and it probably requires an ethnographic, or similar, method, which places high demands on access to the auditors' work and documentation of the same. Properly performed, however, such a study can make valuable contributions to the continued study of the role of the audit in the construction of accounting reliability. Does a certain view of the auditor's role and what is considered as acceptable negotiation strategies lead to either a stronger focus on accounting reliability as a system and initial production of accounting figures or to a stronger focus on accounting reliability in terms of the company's assessment of the business reliability of the figures, or to a stronger focus on how accounting figures are made reliable in order to be communicated to investors and an interested public?

The project’s international dimensions and dissemination

The main means of spreading the research findings described here has (in addition to the publication of articles in scientific journals) been through participation and presentation at international conferences, both nationally and internationally. In this way, the project has been represented at six different conferences and more specialized workshops, as shown in the publication list below. In addition, Gunilla Eklöv Alander was invited to Linnaeus University in Växjö, where she presented the paper Risk and uncertainty in ordering of worth: The production of accounting profit. We are also working on a book manuscript in Swedish. So far, the results have been spread mainly within the scientific community, but the results have also been discussed in different contexts with a large number of practitioners, mainly auditors. Our intention is that when our texts are published in scientific journals, to write up the results in a more easily accessible form.