As part of its mission to “protect investors, maintain fair, orderly, and efficient markets, and facilitate capital formation” the Securities and Exchange Commission (SEC or Commission) has promulgated myriad rules requiring public companies, investment firms, brokers, and other participants in the capital markets to disclose massive amounts of data. Although virtually all public SEC filings have been available electronically on EDGAR since the mid-1990s, only a few designated portions of selected filings have been filed in an “interactive” format that can be reliably read and manipulated by software.
Interactive data has the potential to greatly enhance the reach and utility of disclosures, both for the public and for government agencies tasked with enforcement. It can also reduce costs for filers and consumers of information. Despite the benefits of interactive data, the SEC has promoted the use of interactive data in fits and starts.
After limited initial forays into interactive data in the early 2000s, the SEC seemed to embrace the approach. In 2009, the Commission adopted an ambitious rule (the Interactive Financial Data Rule) that mandated the submission of financial statement data in eXtensible Business Reporting Language (XBRL). That same year, the SEC also began requiring the use of interactive formats for mutual fund risk/return data, nationally recognized statistical rating organization ratings (NRSRO) data, and Form D notices of exempt offerings. A year later, the 2010 Concept Release on the U.S. Proxy System (Proxy Plumbing Release) solicited views on the extension of data tagging to the proxy process.
The extension never occurred. Moreover, the Commission seemed to lose interest in the use of interactive formats. Promising interactive data initiatives have disappeared with little or no warning. Until 2013, only one new form was required to be tagged and only then at the insistence of Congress. At the same time, the Interactive Financial Data Rule has come under heavy criticism, with filers and some academics raising questions about costs, benefits, and quality of the data. Finally, the SEC’s messaging related to interactive data has been mixed: Chair Christopher Cox embraced the concept and spoke glowingly of its potential, while Chairs Mary Schapiro and Mary Jo White have scarcely mentioned it.
Nonetheless, the SEC may be refocusing on interactive data. In 2013, the agency adopted interactive data filings for Form 13F, a quarterly report of holdings by institutional investment managers. The agency’s proposed regulations implementing crowdfunding and Regulation A+ would require issuers to file information as interactive data. In addition, evidence suggests the SEC has recently increased its internal use of tagged data.
As the SEC continues to refine its interactive data strategy, an examination of the agency’s efforts to date may help illuminate the way forward. Part II of this paper provides a brief overview of interactive data, including technological background, and potential benefits. Part III discusses the SEC’s history with interactive data, including the agency’s specific implementations to date. Part IV discusses the issues and criticism related to the SEC’s interactive data efforts. Part V suggests specific implementation strategies and discusses their likely effect on the utility of proposed new interactive data disclosures such as proxy and crowdfunding data, with an analysis of the costs and benefits of various technical and procedural approaches to interactive data.
II. Interactive Data: Background and History
Much of modern securities law and its underlying economic theory rests upon the notion that disclosure can help mitigate market failures caused by information asymmetry. Interactive data magnifies the utility of disclosure by facilitating broader and deeper analysis of the information disclosed. Standardized public company data available in machine-readable formats can decrease costs for investors, analysts, and regulators. Decreased costs for investors lead to more informed investor decisions and greater investor involvement in governance. Lower costs for analysts can expand the universe of covered issuers. Finally, decreased costs for regulators lead to more effective regulation as resources are freed up from the cumbersome task of data collection to the higher-value task of data analysis and consequent regulatory actions.
A. Technical Background
Structured, machine-readable data is by no means a new concept in information systems. The advent of Extensible Markup Language (XML) in the late 1990s, however, facilitated the approach by providing a simple mechanism for structured data to be embedded within “tags,” essentially pieces of metadata that identify the meaning of the underlying data. Tagged data in a structured format is machine-readable through the use of software. The standardized format also allows disparate systems to communicate with each other in the same “language.” In addition to simply producing and consuming such data, software developers can build tools that add value to the “ecosystem” by allowing the data to be aggregated and analyzed across different groupings and classifications.
XML’s structure is intuitive, comprised of building blocks such as elements, attributes, and lists. For example, the following is a possible XML structure to list company directors:
<director age=“45” id=“1”>John Smith</director>
<director age=“34” id=“2”>Jane White</director>
The tags (<company> and <director>) identify elements of data, which in turn have values (NewCo, John Smith, and Jane White) and attributes (age=“45” and id=“1”). Once a structure of elements and attributes is defined, software programs can easily evaluate whether a given XML document conforms to that structure; if it does, the data can be analyzed immediately or stored in a relational database for later analysis.
Most interactive data implementations are derived from the basic XML standard. For example, XBRL is specific to financial data and includes a highly-detailed taxonomy based on GAAP. The specifications are extremely complex, including multiple “schemas” and more than 17,500 distinct elements. Despite this complexity, XBRL conforms fully to the XML standard and can be read and manipulated by standard XML tools. For some initiatives, interoperability among numerous stakeholders is less important, and the SEC has simply developed its own XML standards.
The SEC has taken varied approaches to interactive data technologies. In some cases, the Commission has left responsibility for creating structured data with the filer. In other cases, it has made available “fill-in forms” that allow the filers to enter data into an online interface; SEC software then converts that data to the structured format and renders it in HTML for viewing. This eliminates problems with inaccurate tagging and proliferation of unnecessary tags. The downside of this approach is that it involves human data-entry, which can be prone to error and cannot be part of an end-to-end automated filing process. The Commission has primarily used fill-in forms for simple, highly structured disclosures that involve no more than a few pages. To date, each time the agency has offered a fill-in form, it has also offered the option of a direct XML filing.
Many current SEC interactive data filings entail a degree of duplication. Filers typically must submit two sets of documents: the human-readable HTML filings and machine-readable XBRL files. This duplication has added cost, created the risk of inconsistencies, and raised concerns about the trustworthiness of interactive data, particularly if unaudited. Inline XML may eventually allay these concerns. Inline XML permits filers to incorporate XML tagging directly into an HTML document, avoiding the need for a separate tagged file and, therefore, the risks of duplication. Although inline XML has its detractors, it is seen by many as the future of XBRL filings. The technology has been successfully implemented in the United Kingdom, and the SEC has recently signaled a heightened interest in exploring its use here.
B. Interactive Data’s Potential Benefits
From its inception, interactive data disclosure has offered the promise of a disclosure regime in which information can be read and analyzed by computer systems, allowing for automation of processes and greater accessibility of data. Automation of processes can reduce costs throughout the disclosure system, including for filers themselves. Greater data accessibility allows the SEC and investors to analyze data with more sophistication and at a lower cost.
i. Interactive Data Reduces Filing Costs
Generally, interactive data filings promise to reduce filing costs through several means. First, they may offer a way to reduce duplicative filings with different government agencies. Second, they may help filers catch errors earlier in the filing process, avoiding costly restatements and refilings. Third, to the extent that filers integrate production of interactive data into their internal systems and processes, they can take advantage of that data to perform their own analyses, giving them a greater understanding of their own finances and operations. Finally, for smaller filers especially, assembling an interactive data filing using XML or a fill-in form can be less costly than assembling a human-readable filing.
ii. Interactive Data Facilitates Regulatory Analysis
Generally, interactive data allows regulatory agencies to analyze filing data with more speed, sophistication, and depth than is possible with paper or HTML filings. This enhanced analysis has a number of benefits. For the SEC, interactive data allows its staff to review more financial filings with the same amount of resources. In addition, it can use aggregate information from filings to inform its regulatory policies and rulemaking. Finally, the Commission can undertake new types of reviews targeting specific enforcement goals.
The availability of interactive financial data should allow the SEC to review financial filings more efficiently. SEC staff can review more interactive data filings with the same amount of resources, ensuring that the agency fulfills its mandatory review obligations such as those applicable to periodic reports. This is especially important given the myriad responsibilities of Division of Corporate Finance staff.
Interactive data can also influence regulatory policy, as empirical analysis of such data can lead to a deeper understanding of the utility of current disclosure and the need for additional disclosure. A recent example occurred in connection with the analysis of offering information under Regulation D. The Division of Economic and Risk Analysis (DERA) examined over 100,000 Form Ds filed since 2009 in an interactive format and produced an analysis that included total market size, cyclicality, issuer type, investor type, use of financial intermediaries, and statistical correlations with public stock market performance. This sophisticated analysis allowed the SEC to make specific, data-driven recommendations in its proposal to amend Regulation D, including additional disclosure requirements for private fund general solicitation materials. In addition, the availability of interactive data arguably increased the accuracy of cost calculations used to comply with the Paperwork Reduction Act and in the Economic Analysis section of the proposal.
Interactive data can also assist with targeting certain filer behaviors for enforcement. SEC officials have suggested that the agency may have used Form 4 interactive data to assist in detecting instances of options backdating. Recently, the Director of DERA indicated that the commission would increase its use of interactive data for enforcement purposes. He described a new program called the Accounting Quality Model that “develops custom analytics . . . to inform monitoring programs.” Specifically, DERA (formerly known as the Division of Risk, Strategy, and Financial Innovation) undertook a project to identify “earnings management,” or discretionary manipulation of earnings data related to accruals, a practice that sometimes constitutes fraud. The data also held the potential for increased detection of other accounting issues such as excessive off-balance sheet transactions.
iii. Interactive Data Empowers Investor Analysis and Participation
Interactive data can heighten the ability of investors to analyze filer data. According to the SEC, interactive data will facilitate “comparison of financial and business performance across companies, reporting periods, and industries[;] . . . increase the speed, accuracy, and usability of financial disclosure[;] . . . [and] allow investors to avoid additional [data access] costs associated with third party sources.” Others have noted that interactive data could be employed in a number of areas beyond financial statements, including continuous disclosure requirements, so-called “Key Performance Indicators,” sustainability metrics, shelf registrations, proxy statements, and asset-based securities.
Interactive issuer data may also increase analyst coverage, particularly for smaller issuers. Market observers estimate that at least 29% of all companies traded on United States markets have no “meaningful” analyst coverage. This so-called analyst “neglect” increases as market capitalization decreases; for companies with market capitalization of less than $250 million, the number increases to 55%. To the extent that interactive data helps analysts perform their research more efficiently, they may be able to allocate their time to covering additional companies.
iv. Interactive Data Enables More Effective Corporate Governance
Interactive data can also improve corporate governance. Mainstream corporate governance theory relies on a shareholder primacy model, in which shareholders, as residual owners of the corporation, oversee the board through their shareholder rights, most notably voting. Yet participation rates of individual shareholders are low, resulting in securities intermediaries voting many “uninstructed” shares according to their own default rules. At the same time, investment funds that hold large blocks of shares may lack the resources to diligently research resolutions and directors that come up for shareholder votes; these funds often rely on so-called proxy advisory firms for voting recommendations.
Interactive data can help shareholders become more informed about compensation and voting issues (and thus more engaged in the voting process), but can also help market observers better analyze the effects of uninstructed voting, fund voting, and proxy advisory recommendations. Interactive proxy statement data might even increase shareholder participation. In a 2009 presentation related to notice, access, and proxy voting participation, Broadridge listed “XBRL-filed proxy statements [and] searchable, comparison-enabled information, with link to vote execution” as a “[w]ay to [i]ncrease [p]articipation with [l]ow or [n]o [c]ost to [i]ssuers.” In other words, once interactive proxy statement interactive data is available, the SEC or third parties can develop searchable online resources that present the universe of available proxy data alongside tools for comparison and analysis, links to related interactive data, and links to the online voting applications themselves. Investors who have ready access to all of this data are, in theory, more likely to vote their shares. Ultimately, then, all of this interactive data will enhance the ability of individual fund investors to analyze both the proposals voted on and the votes themselves; investors will apply pressure to funds to vote with more engagement, leading funds to apply pressure to board and management decisions.
III. The SEC’s History with Interactive Data
The SEC’s first foray into interactive data filings came in 2003. Rule changes required that share ownership forms (Forms 3, 4, and 5) be filed using an online fill-in form or by submitting SEC-defined XML. The ownership forms were short and highly-structured, making it relatively easy for Commission staff to convert them to an interactive format. Although the forms were filed as interactive data, the staff also developed software to automatically render that data in HTML for viewing. They did not, however, build tools or feeds to facilitate aggregation, validation, or analysis of that data.
In 2005, the SEC adopted a voluntary program for XBRL reporting of issuer financial data. Designed to “allow registrants, the Commission and others to test and evaluate tagging technology,” the program provided for filers to attach XBRL filings as new exhibits to financial statements. The Commission promised to monitor the use of XBRL by investors, analysts, and third-party software developers and determine “whether adequate safeguards [were] in place to ensure that the data [was] prepared and disseminated correctly.”
The efforts were supported by the creation of the Office of Interactive Disclosure (OID). The OID received a mandate to coordinate the modernization of disclosure and to work with financial data stakeholders to “advance the use of interactive data in financial reporting.”  It spent much of its first two years preparing the agency’s Interactive Financial Data Proposal, which proposed making the voluntary XBRL reporting program mandatory. The OID would ultimately be subsumed by DERA and renamed the Office of Risk Assessment and Interactive Data (ORAID).
At about the same time, the SEC launched two online tools to demonstrate the analytical capabilities offered by XBRL. The “Interactive Financial Report Viewer” allowed users to analyze the interactive data filings of companies participating in the voluntary program for financial data. The “Executive Compensation Reader” gave users the capability to analyze compensation data from 500 large companies selected by SEC staff. Both tools included complex searching, filtering, and sorting, and were well-received. Observers described them as “practical demonstrations of the uses of XBRL” and a “brilliant publicity move for XBRL.” Today, however, these tools are no longer available on the SEC website, and it is unclear when or why they were removed.
The voluntary interactive financial data program provided the Commission with insight into the problems and benefits of the approach, much of it coming from responses to a questionnaire the Commission sent to thirty-five participants. The voluntary program became mandatory in 2009 when the SEC adopted a formal rule (the Interactive Financial Data Rule). Based in part on the SEC’s review of the voluntary program and validation of the interactive data submitted during that program, the Rule sought to balance issuer costs against the risks of inaccurate data. In order to lessen the financial impact on issuers and provide a window for them to familiarize themselves with the new requirements and technologies, the Commission phased in the requirements over three years; it also provided a liability safe harbor for twenty-four months after the initial XBRL filing, up until October 31, 2014.
The SEC adopted other rules involving interactive data, including a mandate that mutual funds file risk/return data in an interactive format. The SEC also required interactive data filings by nationally recognized statistical rating organizations (NRSROs). Finally, it mandated interactive data filing of Form D notices of exempt offerings. Like the Interactive Financial Data Rule, the risk/return rule included liability safe harbors for a limited time period.
With these rules in place, the SEC seemed ready to extend interactive data filing requirements to the proxy process. In 2010, the Concept Release on the U.S. Proxy System (Proxy Plumbing Release) solicited public comment on a number of proxy plumbing issues, including an entire section devoted to proxy-related interactive data. The Proxy Plumbing Release recommended extending interactive data to several filings: (1) executive and director compensation and shareholder vote data from Schedule 14A of issuer proxy statements, (2) Form N-PX annual proxy voting records for mutual funds, and (3) shareholder voting results listed on issuer Form 8-K.
Between 2010 and 2012, however, the SEC failed to follow up on these proposals. Nor did the Commission announce any major interactive data initiatives. There were several possible reasons for the drop-off in activity. First, interactive data required dedicated resources, something that may have been in shorter supply after the adoption of the Dodd-Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank), which required the SEC to promulgate a number of new rules. Second, the advent of new leadership at the Commission may have resulted in less emphasis on interactive data initiatives. Third, in the face of Interactive Financial Data Rule criticism from industry and academics, especially criticism of data quality and lack of SEC oversight, the agency may have re-examined some of its assumptions regarding interactive data.
Although Dodd-Frank may have absorbed agency resources, the legislation also forced the Commission to reengage with interactive data. The Act required the SEC to promulgate rules concerning resource extraction (the Resource Extraction Rule) and required that filings be in interactive data format. In 2012, the agency adopted a new Form SD and required that the Form include interactive data when filed in connection with resource extraction disclosure.
Pressure for interactive data also came indirectly from another provision in Dodd-Frank. Section 911 of Dodd-Frank mandated the formation of the Investor Advisory Committee (IAC), a group of outside experts tasked with advising the SEC on issues critical to investors. On July 25, 2013, the IAC issued a list of recommendations regarding tagged data. The IAC Interactive Data Proposal enumerated three specific recommendations for the SEC: (1) adoption of a “Culture of Smart Disclosure,” (2) internal development of applications to reduce issuer filing costs, and (3) prioritized tagging requirements for proxy-related filings. Together, these recommendations were intended to enable data consumers “to retrieve information in a cost effective and highly usable fashion[,] . . . facilitate the SEC’s ability to monitor securities markets[,] . . . [and] facilitate investor participation in the governance process.” The Commission has not yet responded to the recommendations.
In the aftermath of the recommendations, the agency initiated several new interactive data programs. The Commission provided that institutional investment managers who are required to disclose their holdings on a Form 13F do so using an interactive format. Likewise, proposed rules governing crowdfunding offerings and offerings under Regulation A+ would require issuers to file data interactively. The Commission has also begun to make greater analytical use of interactive data, using it to analyze Form D filings and identify earnings management in financial statements.
IV. Interactive Data: Promise and Reality
The potential of interactive data to improve analysis and regulation related to disclosures is well documented. The SEC’s grand experiment with interactive data has, however, been a mixed bag. Although a number of filings have been converted to interactive data, the consumption and use of that data appears to be minimal. In part, this may be due to the limited history of financial data available (the Interactive Financial Data Rule was only enacted in 2009 and phased in over three years). There have, however, been many documented issues with the quality of data filed, particularly with respect to financial data. These issues will need to be addressed in order to realize interactive data’s full potential.
Issuers have resisted interactive data on the grounds that it is too costly to implement. For example, comment letters to the Proxy Plumbing Release asserted that the costs incurred to implement tagging are burdensome and far outweigh the benefits. Some issuers have also asserted that the SEC greatly underestimated those costs by as much as half.
No one disputes that issuers do incur costs to tag their financial data, as they must either develop in-house expertise or hire outside resources to set up the initial tagging processes. Once the initial learning curve is overcome, costs decrease dramatically. Many large companies now have internal staff that have years of experience with XBRL, and, as expected, the large accounting firms have all developed expertise in the area as well. Indeed, evidence from the SEC’s voluntary program shows filers reporting an 85% reduction in costs between the first and second filings. Outside observers have also noted that the cost of the first filing was “not particularly high to begin with.”
B. Data Quality
Data quality (i.e., relevance, timeliness, and accuracy) is of paramount importance in the effectiveness of interactive data filings. For interactive financial data, the main concern has been accuracy. Financial analysts and other consumers who use data aggregation services or cull data from HTML filings on their own are reluctant to use XBRL as long as they feel it is less accurate. Unfortunately, even in 2013, observers are still seeing basic errors in the XBRL filings they study. For example, one analyst found that “simple” 10-Q data such as total shares outstanding is often off by orders of magnitude or missing completely from XBRL filings. Another analyst found inconsistencies in the way shareholder equity is reported. Between 2009 and 2011, SEC staff published five sets of observations based on their review of interactive financial statements, noting recurring problems with negative values, unnecessary tag extensions, incorrect data types, and missing elements.
Data quality issues may have arisen from the failure of filers to adopt a sufficiently rigorous process for conforming to the requirements. This was in part encouraged by the Interactive Financial Data Rule. The Rule specifically excluded interactive data from officer certification under Exchange Act Rules 13a-14 and 15d-14. Without a legal requirement (and the corollary risk of penalties) to certify interactive data, officers have much less incentive to implement a rigorous review process for that data.
Quality concerns also arose at least in part because of the absence of sufficient regulatory supervision. The Rule provided no external audit requirement for interactive data. In addition, the SEC itself has provided little oversight. For the first few years of implementation, the Commission provided general guidelines and compliance commentary, but does not appear to have sanctioned filers for failing to comply with interactive data requirements.
Recent events indicate that the SEC may be ready to address these concerns. This past summer the SEC issued a comment letter to Standard Drilling, Inc. that noted, and requested the correction of, XBRL compliance issues. The letter appears to be the first SEC comment letter to admonish a company for XBRL problems. It remains to be seen, however, whether it augurs a new focus on XBRL compliance, or is simply the exception that proves the rule on the SEC’s lack of data quality enforcement.
V. Creating the Culture of Smart Disclosure
Despite its decade-long history with interactive data, the presence of an administrative structure dedicated to interactive data, and continuing pressure from outside entities, the SEC has not provided much recent guidance on its future interactive data strategy. In fact, Chair Mary Jo White declined to mention interactive data in two recent speeches that focused in part on disclosure and “information overload.” Notably, toward the end of her Oct. 15, 2013 speech to the National Association of Corporate Directors, Chair White described the need to “[C]onsider ways we can improve investors’ access to disclosure [and] consider different methods of presenting and delivering information, both through our EDGAR system and other methods” without once mentioning interactive data. The future of the initiative that in 2006 former SEC Chairman Christopher Cox lauded as a “world-changing revolution” remains, only seven years later, in serious doubt.
Although it has taken some recent steps to increase the use of interactive data, the SEC now has the opportunity to truly set the standard for effective use of interactive data by a regulatory agency. A philosophy of “smart disclosure” should entail a concrete vision and plan for implementing interactive data that includes: (1) policies to enforce accuracy of interactive data, (2) true integration of interactive data into the rulemaking process, (3) rules that enable intelligent cross-referencing of interactive data cross filings, (4) flexible processes to tailor implementation choices to each specific filing, (5) a clear message to stakeholders on the importance of interactive data and the agency’s specific plans to embrace it, and (6) strategic rulemaking processes.
A. Adopt Policies to Enforce and Promote Accuracy
Because data quality has been such an issue, the SEC must ensure that filers are complying with interactive data requirements. To this end, the agency should deploy resources to audit the data for compliance and punish filers who do not comply. Depending upon the interactive data technology used, the agency may need to audit both underlying data and tagging, including abuse of extensions. Interactive data facilitates this type of audit because the machine-readable data can be more easily analyzed and compared using computer software. The same types of tools and analysis used for SEC initiatives such as the accounting Quality Model can be deployed to detect data outliers that are likely to be errors.
The SEC should also encourage compliance through bulletins or statements that make compliance issues public. The Divisions of Corporation Finance and Investment Management sometimes provide comments when the internal review of a filing uncovers issues that could be “improved or enhanced.” Nonetheless, the practice is rare. Until recently, the Commission had “not issued even one comment letter on any of the more than 1.4 million [interactive financial data] errors identified.” There is no reason that the agency should not treat interactive data errors in the same manner as other filing issues that are subject to the comment letter process.
The SEC must also ensure adequate enforcement. In some cases, “errors” will amount to deliberate misstatements. In other cases, they will result from a failure to adequately follow applicable requirements. In the former, the failures should be treated like any other disclosure violation, including sanctions under the antifraud provisions. For the former, a reasonable scheme for deterrence might involve an initial hardship extension for problems with an interactive data filing, after which a filing that does not comply with requirements will be deemed late or not filed, with the typical associated penalties.
The SEC must also enforce filings to prevent incomplete data sets. The Commission recently found that at least 9% of companies that make Regulation D offerings do not even file Form D. For useful data analysis, it is vital that data sets be complete (or at least that the level of incompleteness is well-understood). For required filings, the Commission should ensure that it has a straightforward means of detecting incomplete data sets. In the case of optional filings, for statistical sampling purposes, data sets must be large enough and representative enough of the total population to draw statistically significant conclusions.
B. Integrate Interactive Data into Rulemaking
To ensure the long-term success of interactive data, the SEC should integrate it into all rulemaking activities and, ultimately, into all existing filings. The agency should begin any proposal for a new filing with a rebuttable presumption that the format should accommodate interactive data unless the costs significantly outweigh the benefits. Generally, for a new filing, it should be more difficult to argue that the costs outweigh the benefits, as the filer would likely incur similar costs from preparing interactive data filings as it would from preparing an alternative format, such as a human-readable filing. Conversion of existing filings will require more analysis.
For existing filings, the agency should analyze the costs and benefits of several different approaches: (1) Converting the entire filing, including all viable data elements, to an interactive data format; (2) Requiring interactive data for a targeted subset of data elements for which the benefits clearly outweigh the costs; and (3) Deferring any conversion because of minimal expected benefit. As part of this effort, the SEC should prioritize the conversion of filings based on the cost-benefit analysis, taking into account the network effects of converting multiple related filings at the same time. DERA is already well-positioned within the Commission to carry out cost-benefit analyses and insert itself into rulemaking processes across all SEC divisions.
C. Ensure Cross-Referencing of Data Elements
Because interactive data will be spread across a number of different filings, the specifications for that data should allow cross-referencing of key data in those documents. This may be accomplished in several ways. Data elements may be assigned a globally unique identifier (guid), as with the Central Index Key (CIK) used within SEC systems to identify filers or, preferably, the global Legal Entity Identifier (LEI). Alternatively, elements may be assigned a line item number that is unique within a specific context, such as a list of resolutions voted on at a specific company’s specific shareholder meeting; in this case, each filing that needs to be cross-referenced must have that full context available. See Appendix A for an illustration of the two approaches.
Regardless of the exact technical implementation, as the SEC promulgates new interactive data rules, it must ensure that the appropriate data elements are “keyed” using one of the above methods such that they may be cross-referenced. For example, in the context of interactive proxy data, company officers and directors should be keyed so that data consumers can more easily identify their relationships across multiple companies. Similarly, board and shareholder resolutions should be keyed so that data from the proxy statement and voting results can be cross-referenced. The keying of unique data elements allows for a sort of network effect: as more filings become interactive, the information contained in all filings becomes more powerful, allowing interactive data to fulfill its promise and create a truly informed marketplace.
D. Flexible Implementation Strategy
A culture of smart disclosure can also benefit from establishing a standardized approach to selecting the appropriate technology for a given interactive data filing. Interactive data may be implemented as straight XML, inline XML, or fill-in forms. Straight XML defines a file format with tags and data that are purely machine-readable. In contrast, inline XML embeds the machine-readable tags within the human-readable HTML document; a single document is both human and machine-readable.
Importantly, in all cases the end result is interactive data in an XML format; the different technologies for filing, however, impose different costs on filers and the SEC. Broadly speaking, requiring a separate XML filing in addition to an existing filing imposes the highest costs on a filer. Fill-in forms impose the lowest cost on a filer (but a higher cost on the SEC), with inline XML somewhere in between. Separate interactive data filings also increase the chance of inconsistent data between the filings, potentially raising costs for both filers and the SEC. Because of the different cost-benefit profiles of these technology choices, the Commission should carefully select the best technology for a given filing.
i. A Suggested Rubric for Selecting Interactive Data Technologies
Because separate interactive filings impose the highest costs on a filer and increase the chance of data inconsistencies, and because inline XML minimizes some of the problems historically associated with straight XML or XBRL, the SEC should simply no longer consider separate straight XML filings a viable option. In other words, the SEC should never add a duplicative interactive filing as an exhibit to a non-interactive filing as it did with financial statement data. Instead, the SEC’s baseline solution for any interactive data filing should be a single straight XML filing. It can then weigh the costs and benefits of allowing for an inline option or a fill-in form.
An inline XML option may be appropriate for complex forms that are already filed in a human-readable format. In these cases, replacing the human-readable filing with an interactive data filing that is only machine-readable may be unacceptable to the stakeholders, while an additional XML-only filing is likely to be seen as costly and prone to the problems noted above. Furthermore, inline XML may reduce extraneous tagging extensions, at least in the context of XBRL financial data filings. In fact, inline XML solutions have already been implemented elsewhere: the United Kingdom has required financial statement filing using inline XBRL since 2011. Perhaps recognizing both its promise and real-world success, the SEC has recently signaled a possible increased interest in inline XBRL, issuing an RFP for technology solutions for parsing inline XBRL data.
In cases where a filing is simple and highly structured or the population of filers contains a meaningful subset of smaller firms for whom straight XML filing costs would be especially burdensome, the addition of a fill-in form option may be appropriate. Fill-in forms impose a development cost on the SEC, but make filing easy by eliminating the need for filers to understand tagging and XML. A fill-in form, however, should never be the only filing option, as the cost savings for smaller filers should not foreclose the development of sophisticated automated solutions for larger filers.
Fill-in forms have been used by the SEC and agencies in other countries. The SEC has used fill-in forms for Forms 3, 4, 5, and 13F, with no significant problems reported. In contrast, the Israel Securities Authority has used fill-in forms for its financial data filings, with mixed results, including a high level of initial data inaccuracy. The more complex the filing, the more difficult it may be to capture its nuances in an electronic fill-in form, especially if the content, or even presence, of different data elements is discretionary.
ii. Example: Applying the Rubric to Proxy Data
As an example of the above rubric in action, consider how the SEC might implement interactive proxy data as suggested by the Proxy Plumbing Release and IAC Interactive Data Proposal. Together, the two documents essentially recommend that portions of Schedule 14A (specifically compensation data and voting matters), Form 8-K voting results, and all of Form N-PX be converted to interactive data filings. A tailored approach to this conversion would balance the factors discussed above, including filer costs, agency costs, data consumer value, and implementation time.
Schedule 14A is a relatively complex filing that includes some highly structured data and some less-structured narrative data. Tagging the entire filing is a non-trivial task that would affect virtually all public companies. In addition, analysts and investors are familiar with reading a paper or HTML version of the filing on a company-by-company basis.
Using the analysis suggested in the previous subsection, the SEC’s first step would be to define a straight XML schema for Schedule 14A. It would do so by first analyzing the current structure of Schedule 14A. Then, it would determine which data elements were necessary to enable data consumers to usefully analyze the information in the filing. Finally, it would superimpose an XML structure containing the desired data elements, including any groupings, over the filing’s intrinsic structure.
The second step would be to determine whether to employ inline XML. Because of the complexity of Schedule 14A and the large number of filers affected, a complete conversion to a straight XML filing would arguably be overly burdensome to both filers and data consumers, as well as costly to the SEC. Thus, the SEC should strongly consider defining an inline XML specification based on its straight XML specification. One significant advantage of inline XML here is that, rather than requiring inline tagging of the entire Schedule 14A, the SEC could limit the tagging requirement to compensation and voting data, imposing lower costs on filers than would a full conversion to XML.
The third step would be to determine whether to also employ a fill-in form. One drawback here is that the SEC would need to create a fill-in form covering the entirety of Schedule 14A, even though its goal is to capture only the compensation and voting data in interactive format. (Although the agency could in theory create a fill-in form with just compensation and voting data, this would impose the additional cost of the fill-in form on top of the existing HTML filing.) Thus, in this case, the cost of development of the form might outweigh the benefits to the smaller issuers that might use it.
Form 8-K is considerably less complex than Schedule 14A and is highly structured. It does, however, affect all public company filers. For this reason, in addition to a straight XML specification, the SEC might wish to consider a fill-in form, which would allay the cost concerns of issuers. The SEC would also need to decide whether to convert all Form 8-K filings or just those that address voting results; here the opinions of issuers might be helpful, as it is not clear whether they would prefer a wholesale conversion of all 8-Ks for consistency purposes, or would prefer a more limited scope. This issuer preference could be an important input into the agency’s cost benefit analysis.
Similarly, Form N-PX is relatively simple and highly structured. The Form does not contain significant discretionary material in the form of comments and footnotes. In addition, Form N-PX is filed only by investment funds, and the data contained within the form is structurally quite similar to that contained in Form 13F, which the SEC has already converted to an XML and fill-in interactive filing. Because of its simple structure and low filer impact, Form N-PX is highly amenable to a simple conversion to straight XML, with or without a fill-in form. In fact, it is likely that most funds already have the information systems in place to easily generate human-readable HTML, interactive XML, or indeed any structured format.
E. Communicate a Clear Message on Interactive Data
Once the SEC has clearly defined its internal vision and policies for interactive data, it should clearly communicate them to the public. This would help counteract the confusion that has surrounded the agency’s interactive data efforts over the last few years. In addition, once the agency has prioritized the conversion of existing filings to interactive data, it should consider making that prioritization public. Together, the vision, policies, and prioritization would allow stakeholders to better plan for future interactive data efforts. For example, software companies would be able to determine which technology platforms to use, as well as what types of tools to develop. Similarly, academics might be able to plan for certain types of research if they could better predict the timing and content of future interactive data filings. Perhaps most importantly, communication of a comprehensive vision and plan would restore stakeholders’ confidence in the SEC’s leadership and competence with respect to an enhanced disclosure regime that fully leverages modern technological capabilities.
F. Approach Rulemaking Strategically
Broadened interactive data efforts by the SEC could conceivably affect a large number of filings by issuers, funds, brokers, and investment advisers. For efficiency reasons, the Commission may not wish to employ full rulemaking for every interactive data initiative. At the same time, the SEC must weigh the ability of new interactive data initiatives to withstand public scrutiny and court challenges.
SEC rulemaking is primarily governed by the Administrative Procedures Act (APA), which requires notice and comment for substantive rulemaking. There are, however, exceptions for interpretive and procedural rules. In general, an agency may wish to avoid notice and comment for rulemaking that is time-sensitive or limited in scope and impact (and thus less likely to be challenged). The costs of preparing a full notice, waiting for the comment period, and responding adequately to comments in the final rule may be substantial. For this reason, the agency may choose to treat a change as interpretive or procedural.
A rule may be procedural if it is merely derivative or mechanical in nature. An interactive data rule that only changes the format of the submission of data, not the substance of the data required, would seem to fall into the category of a derivative or mechanical burden, and thus be considered procedural. Technology choices may even help determine how substantive the rules are: for example, an entirely new—and additive—XBRL filing might be considered more substantive than a new requirement for inline XBRL in a current HTML filing.
The SEC’s conversion of Form 13F filings to an interactive data format was arguably procedural. The notice couched the change as a “moderniz[ation]” of the submission, and the notice is careful to note that “Form 13F and its instructions have not been amended.” In fact, no actual agency rules were amended as part of the conversion, just technical specifications. This approach may be viable where the scope of the change is limited and the affected parties seem unlikely to challenge the action, as with Form N-PX.
Rules requiring interactive proxy data filings might also be considered interpretative. Although the process of distinguishing interpretative rules from substantive and procedural rules has been described as “‘fuzzy,’ ‘tenuous,’ ‘blurred,’ ‘baffling,’ and ‘enshrouded in considerable smog’,” many courts consider “rules or statements issued by an agency to advise the public of the agency’s construction of the statutes and rules which it administers” to be interpretative. Arguably, interactive data rules for existing filings merely interpret the requirements of existing statutes and rules.
Ultimately, the SEC has the opportunity to be strategic in its approach to interactive data implementations, only employing full rulemaking under certain circumstances. Although such a decision will necessarily be very fact-specific, the SEC can likely predict which potential rules will receive more scrutiny from powerful lobbying groups such as the Business Roundtable and Chamber of Commerce. This in turn will allow the agency to determine the likelihood and nature of challenges, and tailor its rulemaking approach accordingly.
In an era when information collection and analysis is ubiquitous, and “big data” is now a household phrase, it is long past time for the SEC to fully embrace interactive data. Although the SEC has made great strides toward implementing interactive data and enhancing the opportunities for the public to analyze that data, much work remains, and the agency is at a crossroads of sorts. If the SEC promotes a muddled or incomplete vision (or no vision at all) for interactive data, it will miss the opportunity to lead the way for other regulatory agencies and sovereigns to realize the full potential of interactive data disclosures. In contrast, by establishing clear policies and procedures for integrating interactive data into its disclosure-related activities, the SEC can achieve the culture of smart disclosure and finally begin to realize the vision it laid out nearly a decade ago: to empower regulators, investors, academics, and other market participants and observers by dramatically increasing the use of interactive data.
Appendix A: Comparison of Technical Cross-Referencing Approaches for Interactive Data
Cross-referencing of interactive data may be accomplished by globally unique keys or context-dependent unique line items.
For example, two separate documents could use cross-referencing by unique key as follows (note that these “filings” are both highly stylized and simplified for illustrative purposes):
Document 1 (e.g., proxy data for issuer 1234567)
<resolution guid=“2345678”>Do no evil.</resolution>
<resolution guid=“3456789”>World domination.</resolution>
Document 2 (e.g., mutual fund voting data for Fund 13579)
<vote resolution_guid=“2345678” shares=“10000”/>
<vote resolution_guid=“3456789” shares=“50000”/>
An analyst with both sets of data could easily match the votes in Document 2 to the resolutions in Document 1 by directly matching the “guids” in the appropriate tagged data.
Using line items within context is only slightly more complicated:
Document 1 (e.g., proxy data for issuer 1234567)
<resolution line_item=“1”>Do no evil.</resolution>
<resolution line_item=“2”>World domination.</resolution>
Document 2 (e.g., proxy data for issuer 9876543)
<resolution line_item=“1”>Elect director Thomas Pynchon.</resolution>
<resolution line_item=“2”>Elect director William Gaddis.</resolution>
Document 2 (e.g., mutual fund voting data for Fund 13579)
<vote resolution_line_item=“1” shares=“10000”/>
<vote resolution_line_item=“2” shares=“50000”/>
<issuer cik= “9876543”>
<vote resolution_line_item=“1” shares=“20000”/>
<vote resolution_line_item=“2” shares=“90000”/>
The difference here is that the combination of the resolution_line_item, meeting_date, and issuer_cik is necessary in order to match the votes to the original resolutions. This makes the software code slightly more complex (but not prohibitively so), but removes the need to generate globally-unique keys for every piece of key data.
. A full list of disclosure forms is available online from the SEC. Securities and Exchange Commission Forms List, SEC, http://www.sec.gov/about/forms/secforms.htm (last modified Jan. 17, 2014). The SEC lists over 120 different filings on this page. The Division of Corporate Finance lists over 40 rules and regulations that “apply to disclosure documents filed with the SEC that are subject to review by the Division.” Rules, Regulations and Schedules, Div. of Corp. Fin., SEC, http://www.sec.gov/divisions/corpfin/ecfrlinks.shtml (last modified Dec. 12, 2012).
. EDGAR is an acronym for “Electronic Data Gathering, Analysis, and Retrieval System.” It “performs automated collection, validation, indexing, acceptance, and forwarding of submissions by companies and others who are required by law to file forms with the U.S. Securities and Exchange Commission (SEC).” Important Information About EDGAR, SEC, http://www.sec.gov/edgar/aboutedgar.htm (last modified Feb. 16, 2010). The page also notes that “[c]ompanies were phased in to EDGAR filing over a three-year period, ending May 6, 1996.” Id.
. The terms “interactive data,” “data tagging,” and “machine-readable data” are often used interchangeably. This paper will use the term “interactive data” to refer to data tagging that results in interactive and machine readable data. XML is the most common language used for tagged structured data; eXtensible Business Reporting Language (XBRL) is a specific implementation of XML.
. See, e.g., SEC, Toward Greater Transparency: Modernizing the Securities and Exchange Commission’s Disclosure System (Jan. 2009) (providing a brief history of the SEC’s interactive data initiatives and recommendations for a “Modernized Disclosure System”), available at http://www.sec.gov/spotlight/disclosureinitiative/report.pdf; Letter from Tamara J. Wiseman, Executive Secretary (Mar. 18, 2004) (FDIC FIL-30-2004) (adopting eXtensible Business Reporting Language (XBRL) standards for Call Report data), available at http://www.fdic.gov/news/news/inactivefinancial/2004/fil3004.html; J. Robert Brown Jr., The SEC and Interactive Data: An Example in Practice, The Race to the Bottom (Oct. 17, 2013, 6:00 AM), http://www.theracetothebottom.org/home/the-sec-and-interactive-data-an-example-in-practice.html (discussing an analysis of unregistered offerings by the SEC’s Division of Economic and Risk Analysis; the Division performed the analysis using interactive data from 110,000 online Form D filings). The SEC continues to signal at least some commitment to better availability and use of data and analytics through initiatives such as its Market Structure and Data Website. Market Structure, SEC, http://www.sec.gov/marketstructure/ (last modified Jan. 10, 2014).
. See Interactive Data to Improve Financial Reporting, Securities Act Release No. 9002A, Exchange Act Release No. 59,324A, Investment Company Act Release No. 28,609A, 74 Fed. Reg. 6776, 6789 (Feb. 10, 2009) [hereinafter Interactive Financial Data Rule] (noting the XBRL filing requirements extend to Forms 10-Q, 10-K, 20-F, 40-F, 8-K, and 6-K); Interactive Data to Improve Financial Reporting, 73 Fed. Reg. 32,794 (June 10, 2008) [hereinafter Interactive Financial Data Proposal].
. See Interactive Data for Mutual Fund Risk/Return Summary, 74 Fed. Reg. 7748 (Feb. 19, 2009) [hereinafter Interactive Risk/Return Data Rule]; Amendments to Rules for Nationally Recognized Statistical Rating Organizations, 74 Fed. Reg. 63,832 (Dec. 4, 2009) [hereinafter Interactive NRSRO Data Rule]; Division of Corporation Finance Guidance on Form D Filing Process, SEC, http://www.sec.gov/divisions/corpfin/formdfiling.htm (last visited Oct. 15, 2013).
. For example, the SEC created an internal database that tagged compensation data for a subset of companies and opened it up for public use; the project was implemented, publicized, and then, despite positive public reaction, abruptly removed. See Press Release, SEC, Chairman Cox Unveils New Internet Tool with Instant Comparisons of Executive Pay (Dec. 21, 2007), available at http://www.sec.gov/news/press/2007/2007-268.htm; Karey Wutkowski, UPDATE 2-US SEC Launches Web Tool to Compare Executive Pay, Reuters (Dec. 21, 2007), http://www.reuters.com/article/2007/12/21/sec-paydata-idUSN2132115320071221; Bob Schneider, The New SEC Reader for Executive Compensation, Hitachi Data Interactive (Jan. 5, 2008), http://web.archive.org/web/20090106105358/http://hitachidatainteractive.com/2008/01/05/the-new-sec-reader-for-executive-compensation/.
. See Trevor S. Harris & Suzanne Morsfield, An Evaluation of the Current State and Future of XBRL and Interactive Data for Investors and Analysts (2012), available at http://www4.gsb.columbia.edu/filemgr?&file_id=7313156 [hereinafter Columbia White Paper]. See also Letter from Douglas K. Chia, Assistant Gen. Counsel of Johnson & Johnson (Oct. 19, 2010), available at http://www.sec.gov/comments/s7-14-10/s71410-114.pdf; Letter from Jane Sherburne, Senior Exec. Vice President (Oct. 19, 2010), available at http://www.sec.gov/comments/s7-14-10/s71410-128.pdf.
. See, e.g., Christopher Cox, SEC Chairman, Opening Statement—Open Meeting on the Use of Technology to Improve Financial Reporting (May 14, 2008), available at http://www.sec.gov/news/speech/2008/spch051408cc.htm. Cox was Chair of the SEC from 2005–2009; Schapiro was Chair of the SEC from 2009–2012; White has been Chair of the SEC since April 2013.
. Crowdfunding, 78 Fed. Reg. 66,427, 66,524 (Nov. 5, 2013); Proposed Rule Amendments for Small and Additional Issues Exemptions Under Section 3(b) of the Securities Act, 79 Fed. Reg. 3925 (Dec. 18, 2013).
. See Brown, supra note 5 (discussing the SEC’s analysis of Form D filings in connection with rulemaking); Craig M. Lewis, Chief Economist, SEC, Risk Modeling at the SEC: The Accounting Quality Model (Dec. 13, 2012), available at https://www.sec.gov/News/Speech/Detail/Speech/1365171491988 (discussing the SEC’s use of interactive financial data to detect so-called “earnings management” by filers as part of the “Accounting Quality Model” initiative).
. One need only examine the quantity of SEC rules related to filings and disclosures to see this theory in action. See, e.g., Cary Coglianese, Richard Zeckhauser & Edward Parson, Seeking Truth for Power: Informational Strategy and Regulatory Policymaking, 89 Minn. L. Rev. 277, 281–82 (2004) (“Government regulation is usually justified on the basis of three main types of market failures: lack of competition . . . [,] externalities . . . [,] and a lack of full information . . . .”); Stephen M. Bainbridge, Mandatory Disclosure: A Behavioral Analysis, 68 U. Cin. L. Rev. 1023, 1024, 1030–34 (2000) (noting “the centrality of mandatory disclosure to the U.S. securities regulation scheme” and discussing rationales for mandatory disclosure based on economic principles of public goods, externalities, and information asymmetry).
. Indeed, a number of companies have built entire product lines on their ability to take the non-standardized data provided by public companies, “cleanse” and standardize it, and resell it to eager consumers. Companies that perform these services include Bloomberg, CapitalIQ, Thomson Reuters, FactSet, and Hoovers. Data tagging can make the information available to investors and market participants without the need for intermediaries.
. A trivial example of such tagging is: <profit>1000000</profit>. The tag “profit” is considered metadata (that is, data that describes underlying data), while the number “1000000” represents the underlying data. The combination of metadata and data in a predictable structured format allows computer systems to read and manipulate information.
. See Introduction to XML, w3schools.com, http://www.w3schools.com/xml/xml_whatis.asp (last visited Feb. 8, 2014); How Can XML be Used?, w3schools.com, http://www.w3schools.com/xml/xml_usedfor.asp (last visited Feb. 8, 2014); XML Tree, w3schools.com, http://www.w3schools.com/xml/xml_tree.asp (last visited Feb. 8, 2014).
. See 2013 US GAAP Financial Reporting Taxonomy, xbrl.us, http://xbrl.us/taxonomies/Pages/US-GAAP2013.aspx (last visited Feb. 8, 2014); Fact Sheets, xbrl.us, http://xbrl.us/Learn/Pages/FactSheet.aspx (last visited Feb. 8, 2014); Fin. Accounting Standards Bd., FASB US GAAP Financial Reporting Taxonomy Technical Guide (Jan. 31, 2013), available at http://www.fasb.org/cs/ContentServer?c=Document_C&pagename=FASB%2FDocument_C%2FDocumentPage&cid=1176160594978.
. This was the case with beneficial ownership reports (Forms 3, 4, & 5), reports under Regulation D (Form D), and institutional investment manager holdings reports (Form 13F). See EDGAR Ownership XML Technical Specification (Version 5), SEC, http://www.sec.gov/info/edgar/ownershipxmltechspec.htm (last modified July 6, 2012); Division of Corporation Finance Guidance on Form D Filing Process, SEC, https://www.sec.gov/divisions/corpfin/formdfiling.htm (last modified June 6, 2013); EDGAR Form 13F XML Technical Specification (Version 1.1), SEC, http://www.sec.gov/info/edgar/specifications/form13fxmltechspec.htm (Sept. 23, 2013).
. These “fill-in forms” are online forms that allow users to enter data into fields on a web page, similar to what taxpayers see when they use a program such as TurboTax. The fields are stored as XML, and the XML is used to render the human-readable HTML that appears in EDGAR.
. See, e.g., EDGAR Ownership XML Technical Specification (Version 5), SEC, http://www.sec.gov/info/edgar/ownershipxmltechspec.htm (last modified July 6, 2012); EDGAR Form 13F XML Technical Specification (Version 1.1), SEC (Sept. 23, 2013), http://www.sec.gov/info/edgar/specifications/form13fxmltechspec.htm.
. For example, the Interactive Financial Data Rule, the Interactive Risk/Return Data Rule, and the Interactive NRSRO Data Rule all specify that filers submit interactive data as additional exhibits to existing filings.
Because we believe that the various electronic formats have uses for which each is best suited, we will continue to require the existing ASCII and HTML electronic formats now used in filings. We also believe it is necessary to monitor the usefulness of interactive data reporting to investors and the cost and ease of providing interactive data before we consider discontinuing the use of ASCII and HTML formats and the integration of formats.
Interactive Financial Data Rule, supra note 6, at 6783.
. See What is Inline XBRL and How Will It Affect You?, WDesk.com (Nov. 13, 2013), https://www.webfilings.com/resources/blog/what-inline-xbrl-and-how-will-it-affect-you (describing the SEC’s recent interest in inline XBRL). Although most industry commentary and news articles refer specifically to inline XBRL, I refer here to the more generalized technology of inline XML, which is the same concept applied to any XML schema, not just those promulgated by the XBRL standards-making bodies.
. The primary issues with inline XBRL are: (1) it is more complex to generate than XBRL or HTML alone; (2) third-party vendors who currently work with XBRL will need to adjust to using inline XBRL; and (3) the new HTML 5 standard (which is currently in development) would not, as currently specified, be compatible with inline XBRL. See Charles Hoffman, High Probability US SEC Will Move to Inline XBRL?, Digital Fin. Reporting (Jan. 11, 2011, 7:47 AM), http://xbrl.squarespace.com/journal/2011/1/11/high-proability-us-sec-will-move-to-inline-xbrl.html. The Director of DERA, Craig M. Lewis, remarked early this year that “[T]he real solution to [data quality] is inline XBRL . . . . This seems to be where the industry is moving, and I fully support that.” Q&A with an Expert: The SEC is Developing Tools That Use XBRL Data to Discover Accounting Anomalies and Improve Financial Disclosures, Dimensions (Apr. 9, 2013), http://merrillcompliancesolutions.wordpress.com/2013/04/09/qa-with-an-expert-the-sec-is-developing-tools-that-use-xbrl-data-to-discover-accounting-anomalies-and-improve-financial-disclosures/.
. See HM Revenue & Customs, XBRL Guide for UK Businesses (Apr. 2011), available at http://www.hmrc.gov.uk/ct/ct-online/file-return/xbrl-guide.pdf; WDesk.com, supra note 28. It bears mentioning that XML is not the only structured data language available, but its popularity and the ready availability of tools for viewing and manipulating it put it at the forefront of available languages, and the SEC has already used it fairly extensively. Data may also be unstructured, and some commenters have questioned the need for structured data at all, instead asserting that modern search algorithms can sift through human-readable HTML documents and make intelligent inferences about data elements, producing results that, while not 100% accurate, are good enough for at least some types of analysis. See Columbia White Paper, supra note 11, at 14. This assertion seems to be rebutted, however, by numerous complaints about the data quality and error rates of third party data aggregators.
. See Advisory Committee on Smaller Public Companies, Securities Act Release No. 8666, Exchange Act Release No. 53,385, 71 Fed. Reg. 11,090, 11,117 (Mar. 3, 2006) (recommending “[formation of] a task force . . . to discuss ways to reduce inefficiencies associated with SEC and other governmental filings” and noting “the introduction of XBRL may make this recommendation a more attractive option.”). In the United Kingdom, for a company with “straightforward financial affairs,” a single XBRL filing serves as both a financial data filing and tax return. See HM Revenue & Customs, XBRL Guide for UK Businesses 5 (Apr. 2011), available at http://www.hmrc.gov.uk/ct/ct-online/file-return/xbrl-guide.pdf.
. Advisory Committee on Smaller Public Companies, supra note 32, at 11,118 (noting that “[m]any companies, but especially smaller public companies, find the EDGAR system unnecessarily . . . costly” and encouraging “the Commission to pursue the use of Internet standards [such as] XBRL . . . to reduce costs”). Note that the promise of cost reductions was a key part of the recommendations that led to the 2009 mandate. See id. at 11,117 (recommending “[formation of] a task force . . . to discuss ways to reduce inefficiencies associated with SEC and other governmental filings” and noting “the introduction of XBRL may make this recommendation a more attractive option”).
. The Sarbanes-Oxley Act in particular requires an “[e]nhanced review of periodic disclosures by issuers”; the SEC must review the financial statements of issuers of exchange-traded public securities no less frequently than once every three years. See 15 U.S.C. § 7266 (2012). The SEC reports on this requirement in its annual report; since 2007 it has met or exceeded its target of reviewing 33% of issuers every year, with a rate of 48% in 2011. SEC, FY 2011 Performance and Accountability Report 42 (Nov. 15, 2011), available at http://www.sec.gov/about/secpar/secpar2011.pdf#performancesummary. This rate could increase further as interactive data takes greater hold.
. The Division not only reviews financial statements, but reviews registration statements, responds to no action requests, engages in rulemaking, reviews tender offers, provides interpretive guidance, and handles issuers’ requests for confidential treatment of filings. See SEC, Assessment of Corporation Finance’s Confidential Treatment Processes and Procedures ii (Sept. 28, 2010), available at http://www.sec.gov/about/offices/oig/reports/audits/2010/479.pdf.
. Vladimir Ivanov & Scott Bauguess, SEC, Capital Raising in the U.S.: An Analysis of Unregistered Offerings Using the Regulation D Exemption (July 2013), available at http://www.sec.gov/divisions/riskfin/whitepapers/dera-unregistered-offerings-reg-d.pdf. The study began with 2009 data because that was when the SEC began requiring interactive Form D filings.
. Amendments to Regulation D, Form D and Rule 156, Securities Act Release No. 9416, Exchange Act Release No. 69,960, Investment Company Act Release No. 30,595, 78 Fed. Reg. 44,806, 44,821–28 (July 24, 2013).
. See Stock Options Backdating: Hearing Before the S. Comm. on Banking, Housing, & Urban Affairs, 109th Cong. (2006), available at http://www.gpo.gov/fdsys/pkg/CHRG-109shrg50300/html/CHRG-109shrg50300.htm; Susan M. Brunka, Xbrl and the Sec: How the Commission Uses Interactive Data to Investigate Illegal Stock Options Backdating and What Interactive Data Means for the Future of Federal Securities Law Enforcement, 44 Cal. W. L. Rev. 87, 108 (2007).
. Craig M. Lewis, Chief Economist, SEC, Risk Modeling at the SEC: The Accounting Quality Model (Dec. 13, 2012), available at https://www.sec.gov/News/Speech/Detail/Speech/1365171491988. At the time of his speech, Mr. Lewis was Chief Economist and Director of the Division of Risk Strategy and Financial Innovation; the Division has since been renamed. Press Release, SEC, SEC Renames Division Focusing on Economic and Risk Analysis (June 6, 2013), available at http://www.sec.gov/News/PressRelease/Detail/PressRelease/1365171575272.
. Keating Invs., Analyzing the Analysts: A Survey of the State of Wall Street Equity Research 10 Years After the Global Settlement 1 (Jan. 2013), available at http://keatingcapital.com/wp-content/uploads/2013/01/Analyzing-the-Analysts-01-22-13-FINAL.pdf. Analyst company Zacks estimates the number at 65%, but does not provide its methodology. Education: Aggressive Growth Investing, Zacks, http://www.zacks.com/education/articles.php?id=57 (last visited Jan. 16, 2014). Worldwide, the share of “neglected” public companies is estimated at 35–40%. Justin Canivet, Small Cap Analyst Coverage: An “Under-the-Radar” Dilemma, World Fed’n of Exchanges, http://www.world-exchanges.org/insight/views/small-cap-analyst-coverage-under-radar-dilemma (last visited Jan. 16, 2014).
. Analyst coverage is considered a positive characteristic of a stock, as it tends to increase trading volume, liquidity, and accurate price discovery. See Michael J. Brennan & Avanidhar Subrahmanyam, Investment Analysis and Price Formation in Securities Markets, 38 J. Fin. Econ. 361 (July 1995); Paul J. Irvine, The Incremental Impact of Analyst Initiation of Coverage, 9 J. Corp. Fin. 431 (2003). It may also decrease the cost of capital. See François Derrien & Ambrus Kecskés, The Real Effects of Financial Shocks: Evidence from Exogenous Changes in Analyst Coverage, 68 J. Fin. 1407 (Aug. 2013). Coverage may, however, have negative effects, such as “short-termism” and a reduction in innovation. See Jie He & Xuan Tian, The Dark Side of Analyst Coverage: The Case of Innovation (Northwestern Law Searle Ctr. on Law, Regulation, and Econ. Growth, Working Paper No. 23, 2012), available at http://www.law.northwestern.edu/faculty/programs/searlecenter/workingpapers/documents/He_Dark_Side_Analyst_Coverage.pdf.
. The typical individual investor is a beneficial owner of his securities, while the broker is the registered owner. The registered owner has the actual voting rights, and the individual investor has a contractual right to instruct the registered owner how to vote. Generally, for uncontested votes other than director elections, in the absence of any instructions, the registered owner may exercise its own discretion (including not voting at all). See NYSE Rule 452 (2003), available at http://nyserules.nyse.com/nysetools/PlatformViewer.asp?SelectedNode=chp_1_2&manual=/nyse/rules/nyse-rules/.
. See Proxy Plumbing Release, supra note 8, at 42,989. The SEC recently held a roundtable to discuss proxy advisory firms; participants raised concerns including the firms’ degree of influence, perceived lack of diligence, perceived conflicts of interest, and lack of competition. See Proxy Advisory Services Roundtable Thursday, December 5, 2013, SEC (Dec. 5, 2013), http://www.sec.gov/news/otherwebcasts/2013/proxy-advisory-services-roundtable-120513.shtml.
. For example, tagging of compensation data from Schedule 14A of the proxy statement would facilitate comparisons of that data across companies and help investors better take advantage of say-on-pay and pay ratio disclosures. Proxy Plumbing Release, supra note 8, at 43,006. The SEC proposed pay ratio disclosure rules on September 18, 2013; the proposed rules include disclosure of the pay ratio on Schedule 14A of issuer proxy statements. Pay Ratio Disclosure, Securities Act Release No. 9452, Exchange Act Release No. 70,443, 78 Fed. Reg. 60,559 (Oct. 31, 2013).
Tagging of proxy voting data from Forms 8-K and N-PX would allow investors to better analyze proxy proposal outcomes and assess the voting records of mutual funds, allowing them to put more pressure on funds to be engaged and vote in shareholder’s best interests. See SEC, Recommendations of the Investor Advisory Committee Regarding the SEC and the Need for the Cost Effective Retrieval of Information by Investors 5–6 (July 25, 2013) [hereinafter IAC Interactive Data Proposal], available at http://www.sec.gov/spotlight/investor-advisory-committee-2012/data-tagging-resolution-72513.pdf.
Tagging of Form N-PX data would allow investors to see mutual fund voting patterns “across funds and across time”; further, it would provide a superior alternative to third-party aggregators whose data is costly, incomplete, and lags filings by several months. The current Form N-PX is unstructured, and funds may use different formats as well as switch formats over time, making it difficult for programmers to reliably automate the extraction and analysis of data See Letter from Michael Ostrovsky, Assoc. Professor of Econ., Stanford Law Sch., to Elizabeth Murphy, Sec’y, U.S. Sec. & Exch. Comm’n (Sept. 5, 2013) [hereinafter Ostrovsky Letter], available at http://www.sec.gov/comments/265-28/26528-36.pdf.
. Broadridge Financial Solutions, Inc. is a private company that handles proxy distribution and voting, including e-proxy. See Overview, Broadridge, http://www.broadridge.com/company-information/about/overview (last visited Oct. 15, 2013).
. See Memorandum from Martha Haines, Chief, Office of Mun. Sec., on meeting with representatives of Broadridge Fin. Solutions, Inc.relating to SR-MSRB-2009-02, at 10 (May 15, 2009), available at http://www.sec.gov/comments/sr-msrb-2009-02/msrb200902-6.pdf. Apparently, Broadridge developed an open source taxonomy and donated it to XBRL.us, but any documentation seems to have disappeared. See Bob Schneider, XBRL and the US Proxy System, Hitachi Data Interactive (Aug. 19, 2010), http://web.archive.org/web/20130929215709/http://hitachidatainteractive.com/2010/08/. A search for “proxy statement” on the XBRL.us website does not return any detailed information related to such a taxonomy, but there is a reference to this taxonomy on the “Major Initiatives” page: “Proxy Taxonomy. Broadridge Financial Solutions contributed a proxy taxonomy to XBRL US in Q4 2008.” Major Initiatives, XBRL.us, http://xbrl.us/Learn/Pages/Initiatives.aspx (last visited Oct. 15, 2013).
. Once the data is available and can be aggregated and cross-referenced with other interactive data, technology companies, academics, and agencies will undoubtedly innovate on the tools front; this concept is explored further in Section III.
. Not all scholars believe that shareholders should have greater influence over directors. See Christopher M. Bruner, The Enduring Ambivalence of Corporate Law, 59 Ala. L. Rev. 1385, 1396 (2008) (surveying various approaches to corporate governance theory and specifically comparing “[b]oard-centric [c]orporate [g]overnance” to “[s]hareholder-centric [c]orporate [g]overnance”).
. Mandated Electronic Filing and Web Site Posting for Forms 3, 4 and 5, 68 Fed. Reg. 25,788 (May 13, 2003) (to be codified at 17 C.F.R. pts. 230, 232, 239, 240, 249, 250, 259, 260, 269, & 274). See also EDGAR Ownership XML Technical Specification (Version 5), SEC, http://www.sec.gov/info/edgar/ownershipxmltechspec.htm (last modified July 6, 2012).
. Christopher Cox, Chairman, SEC, Opening Statement—Open Meeting on the Use of Technology to Improve Financial Reporting (May 14, 2008), available at http://www.sec.gov/news/speech/2008/spch051408cc.htm.
. Division of Economic and Risk Analysis Overview, SEC, http://www.sec.gov/divisions/riskfin.shtml (last modified June 18, 2013); Press Release, SEC, SEC Renames Division Focusing on Economic and Risk Analysis (June 6, 2013), available at http://www.sec.gov/News/PressRelease/Detail/PressRelease/1365171575272.
. Press Release, SEC, Chairman Cox Unveils New Internet Tool With Instant Comparisons of Executive Pay (Dec. 21, 2007), available at http://www.sec.gov/news/press/2007/2007-268.htm. Because proxy data is not filed in interactive format, SEC staff apparently manually tagged the data.
. Bob Schneider, The New SEC Reader for Executive Compensation, Hitachi Data Interactive (Jan. 5, 2008) http://web.archive.org/web/20090106105358/http://hitachidatainteractive.com/2008/01/05/the-new-sec-reader-for-executive-compensation/.
. The SEC’s most recent mention of the Executive Compensation Reader was in the Proxy Plumbing Release. Proxy Plumbing Release, supra note 8, at 43,008 (“[T]here was substantial interest in financial Web pages that linked to the Executive Compensation Reader that temporarily was posted on our Web site beginning in late 2007.”) (emphasis added). It is conceivable that the OID did not want to indefinitely allocate resources to manual tagging for the Executive Compensation Reader. The SEC’s most recent mention of the Interactive Financial Report Viewer was a January 2008 speech by the Director of the Division of Corporate Finance. John W. White, Director, Division of Corporation Finance, SEC, Speech by SEC Staff: Corporation Finance in 2008—A Focus on Financial Reporting (Jan. 23, 2008), available at http://www.sec.gov/news/speech/2008/spch012308jww.htm. The Viewer’s source code was apparently provided for download at http://www.sec.gov/spotlight/xbrl/xbrlviewerlicense.htm, but the download link now redirects to a defunct private website: http://www.viewerprototype1.com/downloads/secviewer_src_2008-12-24.zip.
. The voluntary filings revealed potential issues with accurate tagging, unnecessary tagging extensions, data errors, and mathematical errors. Based on its experience with tagging and data problems, the Commission announced that it planned to create automated validation tools to detect these problems. Interactive Financial Data Rule, supra note 6, at 6793.
. The questionnaire revealed that issuers incurred a substantial cost to prepare their first XBRL filings, but saw an 85% reduction afterward. Id. at 6793, 6805. The questionnaire itself is no longer available on the SEC website, but a search of “XBRL Questionnaire” on the SEC’s legacy search page returns a partial cached copy of the text:
Directly Filing Agent Did you or your filing agent create the XBRL files Do you currently map your accounting database with XBRL Manually Software Does your accounting package support XML/XBRL schemas Yes No What software did you use to create your financial statements in XBRL format Did your independent public accounting firm perform attestation procedures as to the XBRL data Overall, was it difficult or easy to create and transmit XBRL data Yes No Were you able to use XBRL data for internal analysis Yes No Do you believe that your investors or analysts would find XBRL data useful Do you plan to continue to submit XBRL data on EDGAR Do extensions benefit or hinder the ability to use XBRL Benefit Hinder Are extensions important to your internal use of XBRL What incentives would you recommend to encourage all filers to adopt XBRL May we contact you to follow up on your responses to this questionnaire[?]
Search SEC Documents (Legacy Search), SEC, http://search.sec.gov/secgov/index.jsp#queryResultsTop (type “XBRL Questionnaire” in the “Search Terms” field and click “Search”) (mistakes in original) (last modified July 30, 2009).
. Division of Corporate Finance Guidance on Form D Filing Process, SEC, http://www.sec.gov/divisions/corpfin/formdfiling.htm (last modified June 6, 2013). In addition to including interactive data requirements in adopted rules, the Commission also included interactive data requirements in some proposed rules during this period. See, e.g., Asset-Backed Securities, Securities Act Release No. 9117, Exchange Act Release No. 61,858, 75 Fed. Reg. 23,327 (May 3, 2010), available at http://www.sec.gov/rules/proposed/2010/33-9117fr.pdf.
. Interactive Risk/Return Data Rule, supra note 7, at 7749 (allowing issuers exemption from liability arising from XBRL filings for twenty-four months after the initial XBRL filing, up until Oct. 31, 2014).
. See Proxy Plumbing Release, supra note 8, at 43,006–08. The Proxy Plumbing Release posed questions such as whether interactive proxy data would be beneficial for investors, what costs would be involved, which types and elements of data should be interactive, and what the costs and benefits of different technology options would be. Id.
. The SEC did issue a rule in 2010 requiring money market funds to file holdings data in XML. See Money Market Fund Reform, Investment Company Act Release No. 29,132, 75 Fed. Reg. 10,060 (Feb. 23, 2010).
. See Kathleen L. Casey, Commissioner, SEC, Speech by SEC Commissioner: The Regulatory Implementation and Implications of Dodd-Frank (Jan. 23, 2011), available at http://www.sec.gov/news/speech/2011/spch012311klc.htm (discussing the effect of Dodd-Frank on the SEC, specifically noting “the SEC has to do nearly six times (95) the rulemakings and nearly three times the number of studies it had to do under Sarbanes–Oxley, most of them within one year.”). The JOBS Act, signed on April 15, 2012, may also have had an effect on SEC resource allocation.
. See, e.g., Emily Chasan, Costly Data Go Untapped, Wall Street Journal: CFO Journal (Jan. 22, 2013, 12:49 AM), http://blogs.wsj.com/cfo/2013/01/22/costly-data-go-untapped/ (asserting that “XBRL lost its priority status at the SEC amid the financial crisis and its aftermath”); Matt Kelly, As Priorities Shift, XBRL Loses Spotlight at SEC, Hitachi Data Interactive (June 1, 2009), http://web.archive.org/web/20110410092221/http://hitachidatainteractive.com/2009/06/01/as-priorities-shift-xbrl-loses-spotlight-at-sec/ (“Schapiro believes the SEC has much, much bigger problems to worry about than XBRL . . . two XBRL projects planned for this year . . . have been put off until next year”).
. Disclosure of Payments by Resource Extraction Issuers, Exchange Act Release No. 67,717, 77 Fed. Reg. 56,365 (Sept. 12, 2012) [hereinafter Resource Extraction Rule], available at http://www.gpo.gov/fdsys/pkg/FR-2012-09-12/pdf/2012-21155.pdf. The DC Circuit invalidated the rule on July 2, 2013. Am. Petroleum Inst. v. SEC, No. 12-1668 (D.D.C. Jul. 2, 2013).
. Resource Extraction Rule, supra note 84, at 18. Interestingly, another SEC rule related to conflict minerals did not require interactive data filings. Conflict Minerals, 77 Fed. Reg. 56,273 (Sept. 12, 2012) (to be codified at 17 C.F.R. pts. 240 & 249) [hereinafter Conflict Mineral Rule]. Although the SEC did not explain why the Conflict Mineral Rule disclosures do not take advantage of interactive data, two possible reasons are: (1) where the Resource Extraction Rule disclosures are financial, the Conflict Mineral Rule disclosures are descriptive, and perhaps less useful in an interactive format; and (2) Congress did not explicitly require interactive data for the latter.
. Letter from Douglas K. Chia, Assistant Gen. Counsel to Johnson & Johnson, to Elizabeth Murphy, SEC Sec’y (Oct. 19, 2010), available at http://www.sec.gov/comments/s7-14-10/s71410-114.pdf (“[T]ime and expense for registrants to comply with [the Interactive Data Financial Rule] have gone far beyond what anyone expected . . . .”).
. See Letter from Katherine K. Combs, Interim CEO & President of the Soc’y of Corporate Secretaries & Governance Prof’ls, to Elizabeth Murphy, SEC Sec’y (Oct. 28, 2010), available at http://www.sec.gov/comments/s7-14-10/s71410-265.pdf. (“[S]urvey data from Society [of Corporate Secretaries & Governance Professionals] members shows that more than twice as much time [as the SEC estimated] was actually required [for Form 10-Q filings].”); Letter from Jane Sherburne, Senior Exec. Vice President & Gen. Counsel of BNY Mellon, to SEC (Oct. 19, 2010), available at http://www.sec.gov/comments/s7-14-10/s71410-128.pdf (“[T]he cost of implementing XBRL has been meaningfully higher than originally estimated.”).
. Michael Alles & Glen L. Gray, A Relative Cost Framework for Rethinking Assurance of XBRL Filings 6 (2011), available at http://eycarat.faculty.ku.edu//myssi/_pdf/2-Alles-Gray-Relative%20Cost%20and%20XBRL%20Assurance.pdf.
. See CFA Institute, CFA Institute Member Survey: XBRL 19 (2011), available at http://www.cfainstitute.org/ethics/Documents/Research%20Topics%20and%20Positions%20Documents/xbrl_member_survey_report_2011.pdf (noting that “reliability” is the most important attribute of financial data among chartered financial analysts).
. Timeliness is no longer a significant issue for interactive financial data, as the grace periods have been exhausted by most issuers, meaning most XBRL filings occur simultaneous to the HTML filings. For those interactive data filings that have supplanted HTML filings, such as Form D and Form 13F, timeliness has actually improved, as data consumers no longer need to wait for a data aggregator or spend their own time manually culling data. Depending on the complexity or variability of a particular filing, it can take months for even a sophisticated data aggregator to prepare data. Ostrovsky Letter, supra note 53 (noting that data aggregators take several months to prepare N-PX voting data after filing). Similarly, relevance has not been a significant issue (and consumers have indicated they would prefer the inclusion of even more data elements). See Columbia White Paper, supra note 11, at 28.
. See, e.g., David Trainer, XBRL Would Be Wonderful If It Always Worked, Forbes, Great Speculations (Nov. 7, 2013, 11:46 AM), http://www.forbes.com/sites/greatspeculations/2013/11/07/xbrl-would-be-wonderful-if-it-always-worked/.
. Charles Hoffman, Seeing Inconsistencies in Reporting Equity in SEC XBRL Financial Filings, Digital Financial Reporting (Dec. 9, 2013, 3:14 PM), http://xbrl.squarespace.com/journal/2013/12/9/seeing-inconsistencies-in-reporting-equity-in-sec-xbrl-finan.html.
. See Staff Observations from the Review of Interactive Data Financial Statements, SEC (Dec. 13, 2011), http://www.sec.gov/spotlight/xbrl/staff-review-observations-121311.shtml; Staff Observations from the Review of Interactive Data Financial Statements, SEC (June 15, 2011), http://www.sec.gov/spotlight/xbrl/staff-review-observations-061511.shtml; Staff Observations from the Review of Interactive Data Financial Statements, SEC (Nov. 1, 2010), http://www.sec.gov/spotlight/xbrl/staff-review-observations-110110.shtml; Staff Observations from the Review of Interactive Data Financial Statements, SEC (Oct. 6, 2009), http://www.sec.gov/spotlight/xbrl/staff-review-observations-100609.shtml.
. A recent letter from the United States House of Representatives, Committee on Oversight and Government Reform, led by Representative Darrell Issa, took issue with the SEC’s oversight and use of interactive data, suggesting that the SEC has failed to ensure the quality of interactive data and failed to use it for internal reviews, instead relying upon “manual review[s]” of printed-out filings and data from commercial databases, even though its own interactive data is superior in quality. Letter from Darrell Issa, Chairman of the Comm. on Oversight and Gov’t Reform, to Mary Jo White, SEC Chair (Sept. 10, 2013), available at http://oversight.house.gov/wp-content/uploads/2013/09/2013-09-10-DEI-to-White-re-Interactive-Data-Rule.pdf. Note that Issa and the Committee have been criticized for overreaching and grandstanding with some critiques. See, e.g., Ari Rabin-Havt, The Benghazi Report Darrell Issa Would Rather You Not See – And The Media Is Helping Him, Media Matters for America (Sept. 27, 2013, 8:27 AM), http://mediamatters.org/blog/2013/09/27/the-benghazi-report-darrell-issa-would-rather-y/196119; David Weigel, Darrell Issa's Big New IRS Revelation About the White House Was Actually Reported Two Months Ago, Slate (July 18, 2013, 2:44 PM), http://www.slate.com/blogs/weigel/2013/07/18/darrell_issa_s_big_irs_revelation_about_the_white_house_was_revealed_two.html.
. Letter from Mark P. Shuman, SEC Branch Chief-Legal, to Adriaan Reinders, Chief Exec. Officer, Standard Drilling, Inc. (July 2, 2013), available at http://www.sec.gov/Archives/edgar/data/1158694/000000000013036225/filename1.pdf (noting that “[t]he XBRL Document and Entity Identification Information rendered as part of your filing appears to contain a number of data element errors” and also requesting that Standard Drilling revise its XBRL to comply with SEC requirements).
. See Rob Blake, SEC XBRL Comment Letter Update: So Close and Yet…, Trintech (Nov. 20, 2013), http://www.trintech.com/cadency-blog/sec-xbrl-comment-letter-update-so-close-and-yet/ (commenting that “if CorpFin doesn’t follow through . . . the overall impact [of SEC comments on XBRL] is watered down”).
. See Mary Jo White, Chair, SEC, The Path Forward on Disclosure (Oct. 15, 2013), available at http://www.sec.gov/News/Speech/Detail/Speech/1370539878806; Mary Jo White, Chair, SEC, The Importance of Independence (Oct. 3, 2013), available at http://www.sec.gov/News/Speech/Detail/Speech/1370539864016.
. The IAC hit upon a number of these recommendations in its Interactive Data Proposal: the suggestion that the SEC create a “culture of smart disclosure” essentially suggests that the SEC embrace interactive data in a more holistic manner than it has to date, through consistent messaging, integration into rulemaking processes, and commitment of agency resources. See IAC Interactive Data Proposal, supra note 53.
. See Q&A with an Expert: The SEC is Developing Tools That Use XBRL Data to Discover Accounting Anomalies and Improve Financial Disclosures, Dimensions (Apr. 9, 2013), http://merrillcompliancesolutions.wordpress.com/2013/04/09/qa-with-an-expert-the-sec-is-developing-tools-that-use-xbrl-data-to-discover-accounting-anomalies-and-improve-financial-disclosures/.
. See Press Release, SEC, SEC Staff to Publicly Release Comment Letters and Responses (June 24, 2004), available at http://www.sec.gov/news/press/2004-89.htm. The SEC and the filer may exchange multiple rounds of comments and responses until the agency is satisfied with the filing.
. Delinquent filings typically constitute violations of the particular act that requires those filings. See, e.g., Gateway Int’l Holdings, Inc., Exchange Act Release No. 53,907, 88 SEC Docket 430, 439 (May 31, 2006), available at http://www.sec.gov/litigation/opinions/2006/34-53907.pdf.
. Specifically, the analysis found that at least 9–11% (and likely significantly more) of companies making Regulation D offerings do not file Form D. See J. Robert Brown Jr., Amending Regulation D: General Solicitations, Empirical Evidence, and Investor Protection (Part 2), The Race to the Bottom (Nov. 7, 2013, 6:00 AM), http://www.theracetothebottom.org/home/amending-regulation-d-general-solicitations-empirical-eviden-1.html. Although Form D is not required, the SEC is authorized to penalize companies for failing to file it.
. Generally, filers must pull data from existing information systems in order to generate a filing. Modern information systems can just as easily produce XML as they can HTML. For example, with respect to Form N-PX, Michael Ostrovsky notes that:
[M]ost mutual funds already have their voting data stored in a structured database form, either internally or on a service provider’s computers, and use computer scripts to produce the N-PX forms that they then submit to the SEC. Modifying these scripts so that they produce output in a different format(a structured, standardized one) is straightforward.
Ostrovsky Letter, supra note 53.
. The global LEI initiative is a multi-country effort to define unique identifiers for legal entities that can be used consistently across systems and jurisdictions. Use of global LEIs in interactive data would thus allow cross-referencing of data between SEC filings and data from other agencies and countries. See Fin. Stability Bd., A Global Legal Entity Identifier for Financial Markets (June 8, 2012), available at http://www.financialstabilityboard.org/publications/r_120608.pdf.
. In addition, the use of open XML specifications (such as XBRL) in combination with keying of unique data could foster the connection of different data taxonomies both within and between agencies and the public. Shared specifications (including both the overall structure and the element definitions) will allow data consumers to more easily aggregate data from disparate sources, enhancing the depth and breadth of their analysis. See Marc van Hilvoorde, XBRL Can Help Regulators Innovate, Hitachi Data Interactive (Sept. 15, 2009), http://web.archive.org/web/20091003070237/http://hitachidatainteractive.com/2009/09/15/xbrl-can-help-regulators-innovate/.
. I use the term “straight XML” to distinguish it from inline XML; in other words, “straight XML” is an XML-only filing, without human-readable elements, such as the XML specification developed for Form 13F and the XBRL specifications used for interactive financial data.
. Note that to date the SEC has consistently complemented its fill-in forms with straight XML filings, allowing smaller filers to take advantage of the simplicity of the fill-in form and larger filers to take advantage of the efficiencies associated with automated processing.
. Inline XBRL allows filers to combined machine-readable tagged data and human-readable markup within one HTML document. See Kathy Hoffelder, New XBRL Version May Answer SEC’s Prayers, CFO (Sept. 18, 2013), http://ww2.cfo.com/gaap-ifrs/2013/09/new-xbrl-version-may-answer-secs-prayers/.
. See Inline XBRL (iXBRL): The Next Generation of SEC Filings, Dimensions (Dec. 18, 2012), http://merrillcompliancesolutions.wordpress.com/2012/12/18/inline-xbrl-ixbrl-the-next-generation-of-sec-filings/ (noting “iXBRL would also deter filers from creating needless tagging extensions in XBRL in an effort to achieve a desired look in the rendered document”).
. Note that the agency had little choice at the time: an inline XBRL taxonomy for financial data had not yet been developed when the SEC implemented the Interactive Financial Data Rule, and fill-in forms for United States financial statements would be extremely complex.
. See HM Revenue & Customs, XBRL guide for UK businesses (Apr. 2011), available at http://www.hmrc.gov.uk/ct/ct-online/file-return/xbrl-guide.pdf. Note, however, that the IFRS standard used in the UK is considered simpler than the GAAP standard in the US, resulting in “less detailed” reporting. See IFRS Overview, NYSSCPA.ORG, http://www.nysscpa.org/ifrs2/overview.htm (last visited Jan. 16, 2014). In addition, the UK’s tagging requirements are less detailed than the SEC’s. See Andrea Whitehouse, End of HMRC’s ‘soft landing’ – No Room for Complacency, CoreFiling Blog (March 26, 2013), http://blogs.corefiling.com/2013/03/end-of-hmrcs-soft-landing-no-room-for-complacency/. The UK, however, has begun to phase in more detailed tagging requirements. See Andrea Whitehouse, Detailed Profit and Loss Tagging – Get Ready!, CoreFiling Blog (Sept. 3, 2013), http://blogs.corefiling.com/2013/09/detailed-profit-and-loss-tagging-get-ready/.
. Tammy Whitehouse, SEC Seeks Contractor to Help Develop Inline XBRL, Compliance Week (Sept. 24, 2013), http://www.complianceweek.com/sec-seeks-contractor-to-help-develop-inline-xbrl/article/313277/.
. The IAC Interactive Data Proposal specifically recommended that the SEC consider developing fill-in forms for filings when feasible, thus obviating the need for filers to deal with the intricacies of tagging. See IAC Interactive Data Proposal, supra note 53.
. Bob Schneider, XBRL: The Israeli Experience, Hitachi Data Interactive (July 30, 2010), http://web.archive.org/web/20101030044447/http://hitachidatainteractive.com/2010/07/30/xbrl-the-israeli-experience/.
. See EDGAR Ownership XML Technical Specification (Version 5), SEC, http://www.sec.gov/info/edgar/ownershipxmltechspec.htm (last modified July 6, 2012); EDGAR Form 13F XML Technical Specification (Version 1.1), SEC (Sept. 23, 2013), http://www.sec.gov/info/edgar/specifications/form13fxmltechspec.htm.
the whole or a part of an agency statement of general or particular applicability and future effect designed to implement, interpret, or prescribe law or policy or describing the organization, procedure, or practice requirements of an agency and includes the approval or prescription for the future of rates, wages, corporate or financial structures or reorganizations thereof, prices, facilities, appliances, services or allowances therefor or of valuations, costs, or accounting, or practices bearing on any of the foregoing[.]
5 U.S.C. § 551 (2012).
[C]ertain . . . principles have emerged to guide the application of the APA's exemption for procedural rules. First, “[i]n determining whether a rule is substantive, [a court] must look at [the rule's] effect on those interests ultimately at stake in the agency proceeding.” Hence, agency rules that impose “derivative,” “incidental,” or “mechanical” burdens upon regulated individuals are considered procedural, rather than substantive. More broadly, the D.C. Circuit has held that “an otherwise-procedural rule does not become a substantive one, for notice-and-comment purposes, simply because it imposes a burden on regulated parties.”
Nat'l Sec. Counselors v. CIA, 931 F. Supp. 2d 77, 107 (D.D.C. 2013) (first alteration not in original) (citations omitted).
. Of course, an issuer might take the position that the addition of even a single tag is a substantive change that amounts to a new piece of information in the filing. Interestingly, the argument could come down to whether metadata is intrinsically substantive or formal. In any case, issuers are much more likely to challenge changes to a ubiquitous filing, such as the proxy statement, than funds are to challenge Form N-PX requirements.
. The Business Roundtable is an association of Chief Executive Officers of United States companies. See About, Bus. Roundtable, http://www.businessroundtable.org/about (last visited March 11, 2014). The United States Chamber of Commerce is an organization that represents United States businesses. See About, U.S. Chamber of Commerce, https://www.uschamber.com/about-us/about-us-chamber (last visited March 11, 2014). Both organizations have been heavily involved in lobbying against regulation and challenging administrative rules, including SEC rules. See, e.g., Bus. Roundtable v. SEC, 647 F.3d 1144 (D.C. Cir. 2011).
. “Big Data” refers to the analysis of very large data sets created by consumers, businesses, and government. See McKinsey & Co., Big data: The next frontier for innovation, competition, and productivity (May 2011), available at http://www.mckinsey.com/insights/business_technology/big_data_the_next_frontier_for_innovation.
* J.D. Student, University of Denver Sturm College of Law. The author would like to thank Professor J. Robert Brown, Jr. of the University of Denver Sturm College of Law, Adam W. Glass of the Data Transparency Coalition, and David Blaszkowsky for their time, effort, and thoughtful feedback.