Materiality in reporting integration in South Africa: A natural language processing analysis

Abstract

Background: The concept of materiality has evolved as companies grapple with non-financial reporting. One result of this is greater diversity in terminology used in reporting narrative. Natural language processing (NLP) analysis offers a promising alternative to traditional content analysis to assess how local reporters converse about sustainability performance, financial performance and integration between the two.

Aim: Integrated and sustainability reports of selected South African companies were assessed to detect which topics and themes reporters viewed as material. Connectivity in terminology was also explored to consider the level of integrated thinking.

Setting: This study was conducted in a market where reporting integration and a dynamic understanding of materiality are well advanced. South Africa is deemed an integrated reporting front-runner.

Method: Natural language processing was applied to 256 reports published by the highest and lowest quality reporters in South Africa as determined by the Ernst & Young Excellence in Integrated Reporting Awards. The NLP source dictionary was compiled with reference to international reporting standards.

Results: Top reporters displayed a more complete and targeted coverage of key sustainability terms and themes. Their reports provided evidence of integrated thinking as seen by their reference to sustainability and financial terms in close proximity.

Conclusion: The evolution of reporting standards and their understanding of materiality, including report target audience and ways of determining materiality, is shaping more strategic report content.

Contribution: New NLP capabilities can be used to assess divergent corporate reporting narratives for exploring their relevance and understanding of materiality.