Back in 2004, the UK’s ICAEW issued a interesting report called Digital reporting: a progress report. In the report they refer to the challenge for XBRL of tagging narrative information. So I’d like to noodle that for a moment.
In addition, the extension of narrative business reporting into areas where the information to be reported is much less amenable to classification by tagging than traditional accounting information, presents a challenge to the concept of reporting enabled or facilitated by technological standards such as XBRL. This will be an increasingly significant handicap as the importance of disclosures in Management’s Discussion and Analysis (MD&A) or the Operating and Financial Review (OFR) grows. This challenge has to be addressed as part of the promotion and development of XBRL. [p.35-36]
In the USA, this comment has relevance to both footnote tagging and corporate action tagging. But I’d like to consider this in the context of sustainability reporting.
The issue with a lot of sustainability data is that it is not performance-related numeric, quantitative data like a financial number – although some is like tonnes of carbon emissions for example. Instead much of it is behaviour-related textual, qualitative data that is evidential in nature. In fact much sustainability reporting could be regarded not as reporting at all but more as the results of an organizational self-assessment survey.
The problem with this kind of data is that even though it can be tagged easily enough, it’s hard to compare – reliably that is. To truly compare two similarly tagged pieces of sustainability evidence almost certainly requires text analysis – something that most (maybe all) traditional corporate reporting software does not offer. So as we move towards a world that ‘integrates’ financial and non-financial data and tries to make connections between the two, we have to face up to the fact that a lot of non-financial data is text and that most reporting tools simply don’t know how to compare and contrast text as opposed to numbers.
