Of the over 400 posts on this blog, only one was devoted to XBRL…

…but it was a relatively popular one, ranking in the top ten per cent by page views. As the post was entitled XBRL: are we missing something?, this level of interest might be taken to indicate that yeah, perhaps we are. In that post, I mentioned how…

• …there was a time (predating the conversion to IFRS) when XBRL was regularly mentioned in speeches by some senior OSC management: in 2007, then-Chair David Wilson envisaged that while XBRL “may not be quite as revolutionary as the Internet has been, it will be so successful that people will wonder how we ever managed without it.” In that same year, the CSA launched a voluntary filing program to “help the Canadian marketplace gain practical knowledge and experience in preparing, filing and using XBRL information” and help “the CSA assess the usefulness of XBRL as it considers whether to make filing in this format a requirement.”

I commented: “This program hasn’t taken off, to say the least: in the last six months as I write this article, only eight documents have been filed under the program, by just five entities.” As of today, participation appears even worse, with no recent filings. Clicking on the link to the CSA’s XBRL website, it’s clear that much of the information (such as the CSA contact people) hasn’t been updated in years. Also, of course, no such Canadian filing requirement ever came to pass. The XBRL Canada website is in similar disrepair, still highlighting a webinar to be held on May 24, 2017. Perhaps ironically though, many or most of Canada’s largest issuers do now file in the US using XBRL, reflecting an SEC requirement for foreign private issuers (effective for fiscal years ending on or after December 15, 2017). And the XBRL International website remains highly active, regularly citing various new milestones and benefits. A recent speech by the FASB Chair was teeming with examples, for instance:

• …our project teams have used XBRL data to understand everything from how companies report on their research and development to the cost of their defined benefit plans.
• We also use XBRL data to help us understand how companies are applying new standards—and identify where we may need to provide support or clarify guidance.
• When the leases standard took effect for public companies in early 2019, we paid close attention to first quarter filings. Specifically, we wanted to see if there were areas in need of more implementation support—from both a standard-setting and a Taxonomy perspective.
• In this case, we use XBRL to gather data about the transparency of a lessee’s operating lease disclosures.  Thanks to XBRL, we’re able to gather detailed data on what publicly traded companies are disclosing about their operating lease liabilities, lease costs, and weighted average discount rates, among other data points.

But, of course, there’s meant to be more to this than helping out standard-setters and regulators. Most of the available commentaries on the broader benefits are generic and theoretical, with little empirical evidence to back up the claims made (and as I wrote in the earlier post, it’s unclear that better transparency and comparability necessarily ultimately lead to better decision-making). Also, consistent with the theme above, much of the key thinking on the issue (at least in Canada) dates back a few years. The reliance of XBRL on an initially laborious tagging process already seems old-fashioned and frankly, unexciting, compared to the possibilities of artificial intelligence. Another recent CPA Canada publication, on Big Data and Artificial Intelligence, muses on how (among many, many other things) “Financial planning and analysis and financial reporting will experience a shift toward higher-skill positions that focus on translating the underlying analytics to support complex business decisions.” It’s often predicted (including on this blog from time to time) that the focus on historical financial statements in their current form will also come to seem outdated, measured against increased capacities for engaging in real-time with the oceans of underlying data: the “Big Data” publication explores several aspects of this. Set against such (as the publication puts it) “unprecedented and unimagined possibilities,” XBRL might hardly seem worth the effort. Ironically perhaps, the XBRL International website often tends to contribute to this impression, devoting much of its space to initiatives and developments that seem to be moving past it. On the other hand, maybe the key point is this:

• Standardized data, like XBRL for financial data, is crucial to building a data ecosystem that can support AI. For an AI system to effectively extract, understand, analyze, and learn from vast quantities of data, requires access to data that is clearly and unambiguously defined. When it comes to financial data, that can only be XBRL.

Anyway, however you look at it, it’s almost certainly to the detriment of Canadian regulators that we’re not further ahead in this – indeed, that there’s never been an organized conversation about it (such as would have resulted, for instance, from a formal regulatory concept paper). Perhaps, in a more XBRL-pervaded environment, we’d be better placed than we are for the next evolutionary step. Or leap…

The opinions expressed are solely those of the author