How does Integrated Report Quality Affect Decision-Making? An equity analyst perspective in the South African market

Abstract

A key source of information about a company is their integrated reports, whose main purpose is to improve the quality of information available to investors. Seminal behavioural finance literature shows that, in situations where market participants perceive potentially imperfect information, they may be inclined to behavioural biases in their estimation of the value of a firm’s shares. However, prior literature also shows that behavioural bias may be present even when there are no evident deficiencies in the information environment. This study explores the relationship between the quality of integrated reports and behavioural bias in the South African equity market. Integrated report quality is approximated by the scores awarded by adjudicators of the EY Excellence in Integrated Reporting Awards. The study focuses on sell-side equity analysts as market participants. A convergent mixed methods design was used. Quantitative analysis was in the form of a structural equation model with latent variables for quality and each of the behavioural biases. The model incorporated data from 2208 forecasts from 316 analysts, across 342 company-years. Semi-structured interviews with 20 sell-side equity analysts were conducted to provide the context necessary for a meaningful evaluation. Across all methods of analysis, findings show that the quality of information in the integrated reports has little to no effect on the analysts’ decision-making processes. Findings from the interviews indicate that analysts consider direct interactions with management to be the most important source of information for their reports and forecasts. In these interactions, they rely on “gut feel” to establish credibility of management’s assertions. Regarding bias, the quantitative analysis showed that herding has a negative and significant association with the quality of analysts’ forecasts. The qualitative analysis confirmed the analysts’ view of the consensus forecast as a benchmark. Considered with findings of the lack of use of integrated reports, it is inferred that analysts view the consensus forecast as a more useful source of information than the information in the integrated reports. Despite the use of indicator variables with prior literature as precedent, the results of the analysis can vary depending on the choice of measurement approximations for the behavioural biases. The study makes a theoretical contribution by connecting research on integrated report quality and behavioural bias. It shows that analyst coverage, even if sponsored by the company, is considered necessary to reduce information asymmetry, despite the production of integrated reports. The study makes a methodological contribution with its analysis of qualitative data from interviews with analysts. Based on the findings, it considers how general debiasing strategies could be used in the context of analysts, thereby making a practical contribution.