Rep. Keith Ellison to SEC: Why No Progress on XBRL?

Learn more about our members’ COVID-19 Projects. Click Here


Rep. Keith Ellison to SEC: Why No Progress on XBRL?

The House Financial Services Committee’s April 29th hearing on the Securities and Exchange Commission’s budget has generated another round of questions on whether and when the agency will resume its stalled transformation from outdated document-based reporting to structured data, fully searchable and open.

Rep. Keith Ellison (D-MN), who was unable to attend the hearing, used his prerogative as a Committee member to submit the following written questions for the record. Rep. Ellison’s questions will be packaged with questions from other members of the Committee and must be answered by the agency.

  • “Is the SEC the only financial regulator that still collects two versions of every financial statement: one in plain text and another in a structured searchable database? If not, which other regulators collect both paper documents and a searchable database.”
  • “How long does the SEC intend to keep this duplicative requirement to file the same information twice?”
  • “In July 2013, the SEC’s Investor Advisory Committee asked the SEC to adopt structured data formats, like XML and XBRL, for everything it collects. Today, most of the SEC’s 800+ forms are just documents, not structured data. When will the SEC respond to the Investor Advisory Committee’s recommendation that the agency adopt structured data formats for its whole reporting regime, to make all of the information fully downloadable and searchable? What will the response be?”
  • “How is the SEC enforcing data quality in XBRL? Has the SEC sent out letters to any firm asking them to fix their data in XBRL? Will the SEC step up efforts to increase compliance?”

Rep. Ellison’s questions illuminate an ongoing failure by the SEC to deliver corporate financial information as reliable, searchable data to investors and markets.

Although it adopted the eXtensible Business Reporting Language (XBRL) structured data format for public companies’ financial statements in 2009, the SEC still collects the information twice: once as a document and once as structured data in XBRL. Moreover, since the agency applies no quality controls to the XBRL version, investors find it difficult to use the data set to make decisions. Finally, the agency has no plan for the data-driven future of its disclosure system.
Aside from corporate financial statements, ownership reports, and transactional feeds from exchanges, almost all the information that the SEC collects is still trapped in old-fashioned documents–rather than being open to investors, markets, and the public as structured data.
Rep. Ellison joins a growing and bipartisan chorus of policymakers, academics, and investors asking the SEC to fix the quality of its existing structured data and expand its use of structured data to include all the information it takes in.

  • In September 2013, Rep. Darrell Issa (R-CA) asked the agency to explain why it hadn’t yet (and still hasn’t) incorporated XBRL into its financial statement review process, nor improved its quality for easier use by investors and markets. The SEC never publicly responded to Rep. Issa.
  • In July 2013, the House Appropriations Committee asked the agency to explain, in a written report, how it would make changes to deliver reliable corporate financial data “in an understandable and accessible format.” The SEC never publicly responded to the committee.
  • In July 2013, as Rep. Ellison’s new questions note, the SEC’s Investor Advisory Committee asked the agency to adopt a “culture of smart disclosure” by integrating structured data into all its reporting requirements. The SEC has not yet responded to the Investor Advisory Committee’s recommendations, though it’s obliged to respond in writing under the Dodd-Frank financial reform law.
  • In January 2013, Columbia Business School’s Center for Excellence in Accounting and Security Analysis reported that investors need structured data on corporate financial performance, but weren’t getting reliable data from XBRL.