What transparency means in the 21st century: A comment and an example
The financial implosion has brought new calls for transparency by businesses and investment funds. The federal stimulus has brought similar calls to track projects and spending openly. The concept of transparency is popular on the left and the right. Who can be against giving taxpayers or shareholders more information? And when the alternative is heavy-handed regulation or opacity, transparency looks like a superior alternative.
What is now clear, though, is that 20th century notions of transparency, defined basically as providing more information, is unlikely to help governments and citizens understand what’s going on in an increasingly complex world. Besides, companies have been spitting out quarterly reports, earnings statements and calls, and accounting records for years, decades in some cases. According to Wired magazine, between 1996 and 2005, the federal government issued more than 30 new financial disclosure rules. The SEC’s public document database adds 200 gigabytes of filings each year, which translates to about 15 million pages of text. That’s up from 35 gigabytes a decade ago.
Lack of data is not the problem. We’re awash in it. There’s so much information that regulators can’t keep up with the constant torrent of it. What’s needed is cleaner, clearer presentations of data, plus more flexible reporting that expands our ability to make use of all the bytes already floating around out there. Harvard’s Kennedy School has launched a center called the Transparency Policy Project with the goal of researching U.S. and international transparency systems in areas like financial reporting, school report cards, toxic pollution reporting and infectious disease surveillance. Both the Transparency Policy Project and Wired argue for releasing packages of data that can be exported to anyone’s web browser or Excel sheet. Just as Wikipedia has empowered ordinary citizens to produce a comprehensive and usually accurate encyclopedia, there seems to be realistic hope that technology can do the same for disclosure reporting of all kinds, allowing individuals to pool their knowledge of auto defects, hospital errors, or disease outbreaks.
An example of this kind of clean, clear, and flexible transparency technology is eXtensible Business Reporting Language, or XBRL for short. XBRL is an open standard for creating electronic reports and exchanging data via the web. Using a standardized series of “tags” for labeling information, XBRL essentially allows anyone to download and analyze huge amounts of data using a simple spreadsheet. By June of this year, companies with a market capitalization over $5 billion that use U.S. accounting rules will need to submit all filings via the XBRL format, according to a recently announced SEC rule. The requirement will be phased in over three years for smaller public companies and mutual funds.
Europe has been an early advocate of mandatory and voluntary XBRL projects, with more than two dozen agencies adopting the technology. Japan has also been a leader. After a small pilot project in 2007 of about 50 firms, more than 1,200 Japanese companies voluntarily enrolled in the second pilot program. The implementation of an industry-wide XBRL rule in 2008 has proceeded smoothly with “few voices of protest or dissatisfaction” according to a key architect of the plan. Similar projects in the U.S. are on a much smaller scale, limited, at the moment, to mostly developmental stages. (Besides the SEC, the Department of Housing and Urban Development has an XBRL project. For a complete map check out XBRLPlanet.)
While opening up databases to professional and amateur analysts everywhere is one strategy for using XBRL, there are others. The technology can be used in a larger effort to streamline the auditing the process and dramatically cut down on the amount of time needed to review a company’s filings. When testing XBRL among a handful of banks, U.S. regulators found that they could review filings in two days – compared with 70 before!
Nowhere has this strategy for streamlining been more explicit than in Holland. The Dutch experience showcases how government can play a productive role in XBRL deployment, primarily through ensuring high-quality standards for data reporting. The combination of intelligent standardization and transparency enhancing technology can actually lead to a reduction in red tape – a goal all public and private parties can applaud.
Started in 2004, the Dutch Taxonomy Project sought to cut administrative burdens by 25 percent over four years (for a savings of more than $500 million each year) through the development of a set of XBRL tags that would function much like a dictionary, and the initiation of XBRL technology. Prior to this project, financial firms submitted some 200,000 different pieces of required information. The Dutch claim to have reduced that number to 8,000 without losing any data. As a point of comparison, the U.S. Generally Accepted Accounting Principles, released in 2007, has about 15,000 terms.
With its taxonomy in place, Holland is moving into a new stage of XBRL reporting. A comprehensive XBRL program in the U.S. may still be years away. The XBRL software market may need to mature, more businesses will need to embrace the idea, and a trimmer set of terms seems necessary. But the technology-taxonomy model is one worth watching, and may eventually be used to sketch in the details of 21st century transparency.