The UK financial sector is a major driver of economic activity and transparent and effective financial communication is a key determinant of its success. Audited financial statements, unaudited corporate disclosures, and information signalled through corporate financial choices are the primary ways that firms communicate with capital market participants. These mechanisms, together with information from analysts, financial journalists, rating agencies and other market commentators external to the firm combine to form the Corporate Financial Information Environment (CFIE).
Narrative disclosures represent a large part of firms’ overall financial communications with investors. Textual commentaries help to clarify issues obscured by complex accounting methods and footnote disclosures. In addition, narratives summarise corporate strategy, contextualize results, explain governance arrangements, describe corporate social responsibility policy, and provide forward looking information for investors.
Our project studied the causes and consequences of corporate disclosure and financial reporting outcomes. The determinants of financial reporting quality and the factors that influence the quality of information disclosed to investors beyond the financial statements were the main focal points of the proposed project.
While a considerable body of research exists on financial narratives, it has been limited by the methods used for measuring the characteristics of such disclosures. In particular, the need to hand-collect relevant data from firms’ annual reports and the subjectivity of textual scoring methods have restricted progress. Recent advances in computing and linguistics provide a basis for undertaking more sophisticated analyses. This project brought together a multidisciplinary team with the aim of developing statistical and computer-based techniques for measuring the properties of UK corporate disclosures. In particular, we have developed new ways of measuring the quality and tone of company narratives using computer-based rankings of annual reports. Both the rankings and linguistic techniques on which these rankings are based have been made available to those seeking information on corporate disclosure policy or wishing to undertake their own analysis of specific narrative statements. We have also used the findings from our analysis as the basis for studying how managers communicate expectations of firm performance to investors and they seek to manipulate investors’ impressions of reported results.
The project has yielded important insights for business policy makers, accounting standard setting bodies and financial market information regulators. We also expect equity market participants including investors, investment analysts, finance directors, auditors, and firm officials to benefit from the research.
The CFIE project ran for just over two years from December 2012. We are now continuing with related financial text analysis work in theUnderstanding Corporate Communications project funded as part of the ESRC Centre for Corpus Approaches to Social Science (CASS). Publications from both projects will continue to be listed here on this website.