Believe it or not, a decade has elapsed since June 2009, when the SEC implemented its XBRL-tagging requirement for financial disclosure filings. Three years later, the XBRL mandate was fully phased in for all SEC filers, and it continues to expand. All regulatory compliance teams at SEC reporting companies are now involved in XBRL tagging. SEC rules that took effect in May 2019 now require Inline XBRL for information on the cover of Forms 8-K, 10-Q, 10-K, 20-F, and 40-F.
To mark the anniversary, Dimensions asked six XBRL experts in the securities regulation, financial reporting, or capital markets sectors to comment on the structured-data revolution in SEC reporting: its benefits to investors and companies; the success stories thus far; and the challenges that remain for structured data and the general modernization of disclosure.
• Mike Willis, Assistant Director, SEC Office of Structured Disclosure
• J. Louis Matherne, Chief of Taxonomy Development, FASB
• Campbell Pryde, President and CEO, XBRL US
• Christine Tan, Co-Founder and Chief Research Officer, idaciti
• Pranav Ghai, CEO, Calcbench
• Lou Rohman, Vice President of XBRL Services, Toppan Merrill
NOTE: The views expressed here are solely those of the individual respondents, and they do not necessarily reflect the views of their respective organizations.
How has the use of XBRL/structured data evolved over the last ten years?
J. Louis Matherne, FASB: That is an interesting question, particularly when you broaden it beyond XBRL to structured data. In the last 10 years, structured data has become almost a household term. It would be vain to think that we
started this movement, but I am confident in saying that we would really be behind the eight ball if we had not started when we did. Big data is all the talk, maybe even superseded by related advances like artificial intelligence, blockchain, and distributed ledger. Imagine where we would be if we had not started when we did?
One general lesson we have all learned about technological advances is that there are generally multiple advances that need to converge before an idea is actually workable—whether we are talking about smartphones or GPS tracking devices, or the amazing number of fitness trackers that are available today and the data they provide. All these technology advances depend on multiple technologies being available and being available at a cost-effective price.
So getting to the question, the broad availability of data—both structured and unstructured, financial and everything else, and the applications and platforms that are now available to consume and use this data—is fundamentally
transforming how analysts and investors consume and use the information. They have multiple sources of information, some to corroborate and some to expand beyond the foundational financial data. Analysts and investors have many sources of information that they consider useful, and they are mashing this together in their earnings models. Much of this was not possible when XBRL was first conceived 20 years ago, but the advance of multiple
technologies has made it a reality.
Lou Rohman, Toppan Merrill: It took investors, analysts, and the SEC a long time to start using XBRL. The data sat idle in the early years. But that period was necessary to get to where it is today, where investors and the SEC are
using it and the efficiencies are being realized. If the SEC had waited for investors to ask for tagged data, we would still be where we were 10 years ago—consuming data from paper-based, non-structured financials.
Even though the data was not used in the first several years, today’s consumers are using data that was filed during the early years. So although no one used the data then, the data from ten years ago is being used today.
Mike Willis, SEC: As with any new standard (e.g., UPC/Bar Code, HTML, etc.), users often consider its usefulness in the context of existing solutions. With the UPC/Bar Code, a common initial reaction was “How does the consumer read these squiggly lines?” in the belief that the grocery-store consumer rather than the scanner would be the primary user. With HTML, the initial reaction included “I can already link documents with my proprietary software; why do I need HTML?” thinking that the limit of linking documents was only for those within their proprietary application. In the case of XBRL and structured disclosures, the initial reaction was to focus on the “comparable” disclosures rather than on the standardization of all disclosures.
As a result, users commonly asked XBRL vendors to replicate their existing data products, which included highly normalized data for a small portion of reported disclosures, rather than looking at the “green field” increase in
disclosed data for 100% of the registrants or the linkages from the disclosures to other relevant resources (e.g., FASB Codification; IFRS Bound Volume).
Next were enhanced visualizations of the more granular structured disclosures. Visualization of detailed disclosures was useful in sorting out proposed tax-policy implications and/or tax-timing differences that may not fully line up
with tax-compliance reports. Structured-disclosure visualizations also included heat maps on tonality and risk of narrative disclosures and variances with numeric disclosure trends. A picture is worth a thousand words, and structured disclosure lowers the reuse burden, including for visualizations enabling more pictures and even videos.
The breadth of XBRL-enabled vendor applications has also continued to expand domestically and internationally. Vendors tuned into the standardized structured disclosures are delivering new capabilities and features that would have been considered somewhat of a “magic trick” a few years ago, including:
- automated identification of mathematical and cross-referencing inconsistencies
- identification of required disclosures, as well as their absence within a report
- benchmarking for all numeric and narrative disclosures
- time series charting and benchmarking for all numeric values
- collaborative/reusable analytical modeling for risk, liquidity, compliance
- aggregation and analysis of detail disclosures embedded within notes
Further, the Inline XBRL format enables these enhanced capabilities and more to be delivered directly “on top of the traditionally formatted report,” sort of a “heads-up display” to aid filers during their drafting and review efforts, and investors and analysts during their assessment efforts.
Campbell Pryde, XBRL US: Initially, large commercial providers were slow to transition to the XBRL filings, despite the fact that XBRL is much easier to process. First, these organizations had a significant investment in their existing
infrastructure and established process, which involved scraping HTML and text filings, reviewing them, and then databasing them. Second, there were quality issues in the early filings. Companies were just getting up the learning curve, understanding how to work with the US GAAP Taxonomy, adapting to new tools. Some companies used an excessive number of extended concepts, which limited the comparability of the reported data, and a lack of explicit
guidance for filers resulted in errors in the XBRL data, such as incorrect signage and mistakes in scaling.
These errors made it difficult to automate the use of the data. But in 2015, we established the XBRL US Center for Data Quality to address these issues. We brought together analysts, XBRL tool and service providers, and corporate data providers to address data quality issues in XBRL filings. Since 2015, we have been creating and making freely available comprehensive, automated rules that issuers use to identify and resolve errors in their filings, before submitting to the SEC.
Through this initiative, we have seen the quality of XBRL data improve significantly. We have also seen that the availability of free, structured data has encouraged new entrants in the data and analytics business. Companies like
Calcbench, Intrinio, Tagnifi, idaciti, and others, really owe their start to XBRL data. With XBRL, data providers no longer need to hire staff to manually vet every reported fact. That process can be largely automated.
We are also seeing that the investment community is using the XBRL data because of the data’s greater granularity, in addition to the fact that it is available faster and is less expensive to process. Plus, the data is available from all filers, large and small, at the same level of detail and timeliness.
Pranav Ghai, Calcbench: The richness of the data is unparalleled. That has led to our clients using footnote-level information in novel ways, including systematic analysis of disclosures and how they change over time (both numbers and text).