As company annual reports get ever larger, so accompanying narratives have grown ever lengthier too. But are the words in danger of clouding the numbers?
Martin Walker, Professor of Finance and Accounting, has been spearheading a major cross-disciplinary study into the nature and impact that narratives have in today’s corporate reporting cycle. Working with colleague Thomas Schleicher, the project was jointly run with Lancaster University and the London School of Economics.
Martin says the central tenet of the work has been to “get on top of” the whole subject of narrative reporting. “Previously the report would essentially have been full of numbers with a brief narrative at the front from the chairman and chief executive. Today you have a whole series of narratives from different players within the business, while the numbers are usually shunted to the back of the report. We set out to determine the quality of narratives, how they differ across companies, how they also change over time, and whether they accurately reflect financial performance.”
Aided by a computational linguist, the team analysed some 12,000 annual reports over a 12 year period, and then extracted some key commonalities.
Says Martin: “One of the most striking observations was how the actual text was pitched. We concluded that the average readability age required to understand these reports was a masters’ level graduate. So immediately there was a certain amount of ‘fog’ around actually understanding what was being said.”
However, the study concluded that firms did benefit by putting a lot of work into their reports. “Ultimately, more detailed reports reflected the willingness of a firm to talk about themselves more in the market and we found that three key things made a difference. Firstly, that forward looking content was very useful to the market; secondly that if the causal language is good and the company can explain what is going on well then that has an impact; and finally that if they explain strategy well that is important too.
“We are now looking to develop further research papers out of the project, such as how the audited accounting numbers affect the narrative and vice versa, and whether poor narratives are ultimately a bad signal for investors.”
He added that the multi-disciplinary nature of the project had also brought significant benefits. “The project combines computer based linguistic analysis, with advanced understanding of accounting and corporate financial communication. As such it is a model of how researchers from two very different disciplines can work together to produce something original and exciting. It was a novel and enjoyable experience for everyone in the team.”