Planning and Evaluation of Assurance Services for Sustainability Reporting: An Evidential Reasoning Approach
ABSTRACT This study develops a framework for planning, performing, and evaluating evidence obtained to assess and control the risks of providing assurance on sustainability reports. Sustainability reporting, or corporate sustainability reporting (CSR), provides stakeholders with important information on both financial and non-financial factors related to environmental, social, and economic performance. Importantly, the presented framework is developed from both a Bayesian (probability-based theory) and Belief Function (Dempster-Shafer theory) perspective. This facilitates application of the framework to cases where the assurance provider prefers to assess risk in terms of probability versus in terms of beliefs. To demonstrate the application of this framework we evaluate assertions, sub-assertions, and audit evidence relevant to CSR based on the G3 Reporting framework developed by the Global Reporting Initiative (GRI). The paper contributes to the literature in three main areas. First, it demonstrates how evidence-based reasoning can be used for engagements where different levels of assurance are provided for the assertions being audited. Second, it shows how various items of evidence at different levels may be aggregated. Third, it presents a generic theoretical model for assuring information based on belief-based assessments, which is then contrasted with a theoretical model based on probability theory. In contrasting the two approaches, we show that in cases where initial uncertainty is substantial, the use of Dempster-Shafer theory has advantages over probability theory in terms of efficiency in achieving a targeted low level of assurance.