- Fund Regulation
- Retirement Security
- Trading & Markets
- Fund Governance
- ICI Comment Letters
August 1, 2008
Florence E. Harmon
U.S. Securities and Exchange Commission
100 F Street, N.E.
Washington, D.C. 20549-9303
Re: Interactive Data for Mutual Fund Risk/Return Summary, File No. S7-12-08
Dear Ms. Harmon:
The Investment Company Institute1 has long favored initiatives designed to tap into the power of technology to better inform mutual fund investors.2 In particular, we have strongly supported the Commission’s proposal to permit open-end management investment companies3 to provide investors with a summary of key information (a “Summary Prospectus”), and make additional information available on the Internet.4 The Summary Prospectus Proposal reflects a strikingly broad consensus that investors would be best served by simplified, streamlined disclosure of essential fund information. It is validated by extensive empirical research conducted by the Commission, the Institute, and others, demonstrating both the preferences of fund investors and their widespread use of the Internet to obtain financial information. We urge the Commission to move forward with that proposal as soon as possible.
We also enthusiastically supported the Commission’s proposal to permit funds to submit the risk/return summary portion of their prospectuses in eXtensible Busin ess Reporting Language (“XBRL”).5 In fact, the Institute coordinated the working group that drafted the taxonomy currently used in the Voluntary Program. We continue to believe that the Commission’s proposal to require funds to provide risk/return summary information in XBRL6 ultimately may further the shared, fundamental goal of better serving the information needs of fund investors. Nevertheless, for the reasons set forth below, we think the mandatory proposal is premature. We strongly urge the Commission to approach this initiative in a methodical, step-by-step manner that will permit certain important elements to fall in place first.
As a preliminary matter, major rule proposals are pending that are likely to result in substantial changes to the risk/return summary.7 As the Commission recognizes, any changes to the risk/return summary as a result of these rule proposals (or for any other reason) would require corresponding changes to the risk/return summary taxonomy.8 For a variety of reasons, tagging risk/return summary information is complicated and time-consuming. It would be extremely inefficient and costly to require funds to begin making XBRL submissions using the existing taxonomy, and then require them to comply with any necessary revisions resulting from these pending rule proposals.
More broadly, while we continue to believe that tagging risk/return summary information may have the potential to benefit investors, we question whether there is sufficient experience to justify requiring it at this time. As we stated in our comment letter supporting the Voluntary Program, the program was an “ideal way to test the use of XBRL in the mutual fund context”(emphasis added).9 The Commission took a similar view, stating that the Voluntary Program was “intended to help us evaluate the usefulness to investors, third-party analysts, registrants, the Commission, and the marketplace of data tagging and, in particular, of tagging mutual fund information.”10 The Voluntary Program has been effective for less than a year.11 More time is necessary to evaluate the current taxonomy, identify all of the technical issues or challenges that may arise, and develop tools that could feasibly enable fund complexes to tag all of their funds.12 Additional time also would enable the SEC and fund companies to assess the true costs and benefits of XBRL for risk/return summary information. Each of these elements must be further explored before the Commission proposes to mandate risk/return summary filings in XBRL.
Finally, certain aspects of the proposal, particularly those addressing funds’ obligations and potential liability, provide cause for concern. These provisions raise significant interpretive questions, and address only a small subset of possible scenarios in which liability could arise. The absence of certain scenarios from the proposal may indicate a need to further consider whether and how the expected uses of XBRL align with the general framework of securities regulation. In addition, the requirement that funds post XBRL files on their websites is unnecessary and may cause investor confusion.
For all of these reasons, we urge the Commission to proceed in a more deliberate manner. First, the Commission should focus its resources on finalizing the Summary Prospectus, which has the potential to benefit shareholders immediately. The Commission should also finalize the ETF proposal, and then turn its attention to revising the risk/return summary taxonomy to correspond to any revisions to Form N-1A. During this time, the Commission should take steps to encourage participation in the Voluntary Program and the development of tools to facilitate XBRL filing for non-financial data. The Commission should repropose mandatory tagging of the risk/return summary when the time is ripe—i.e., after the taxonomy has been revised, has been tested to ensure that it conveys information to investors in the most useful manner, and has been acknowledged, and once the development of tools to facilitate creation and viewing of tagged files is well underway. Finally, we urge that any such reproposal provide clarity with respect to funds’ obligations and potential liability.
These comments are discussed in more detail below.
I. The Commission’s Proposal is Premature
Institute members uniformly believe that other important, pending proposals should be finalized before the Commission requires funds to provide risk/return summary information in XBRL. Under the proposal, the Commission would require approximately 9,500 funds to tag their risk/return summaries, even though the taxonomy upon which all of this rests is likely to be significantly changed—as a result of SEC proposals that predate this one—in the near future. Alternatively, the Commission could attempt both to adopt the proposals affecting Form N-1A and complete the revised taxonomy in time for the effective date of this proposal. In that case, too, the proposal would be premature, because under this approach a realistic compliance date could be two years away or more. While a delayed implementation date is possible, the Commission, the industry, and investors would be better served if the Commission instead took the opportunity to gather more information and refine its proposal.
A. Substantial Changes to the Risk/Return Summary are Pending
Several recent SEC proposals have included changes to the risk/return summary. Most notably, if adopted as proposed, the Summary Prospectus Proposal would substantially revise the order of the information included in the risk/return summary, modify the treatment of fee waivers in the annual fund operating expenses section, and include two new elements: a statement about breakpoint discounts, and portfolio turnover disclosure. The ETF Proposal would also impose a number of changes to the risk/return summary items for ETFs: adding average annual total returns based on market price to the risk/return table; requiring index-based ETFs to include in the table the performance of the index they track rather than a broad-based market index; excluding the fees associated with purchase and redemption of creation units from the fee table; and modifying the fee table narrative to reflect that ETF shares are sold on the secondary market. The Commission also envisions making certain architectural modifications to the risk/return taxonomy to integrate it with the U.S. GAAP taxonomy.13
Requiring funds to utilize the current taxonomy when substantial revisions are likely in the very near term would be extremely inefficient and costly. Once the taxonomy is revised, all of the existing XBRL files will need to be recreated using the new version. As discussed further below, this is a manual and time-consuming endeavor.14 Tools to help with the tagging process will also need to be revised to account for changes to the taxonomy.15 Even once automated tools become available,16 learning a new taxonomy will require an investment of time. Institute members who participated in the voluntary program have reported that there is a steep learning curve in understanding a taxonomy. Significant consideration is necessary to determine how best to apply tags to relevant information, and some trial and error is likely.17 If the Commission were to require compliance with the current taxonomy and then the revised one, which may include a number of new and modified elements, funds would undergo this process twice. Even seemingly simple modifications to the risk/return summary as a result of the Summary Prospectus Proposal, such as changing the order in which items are presented, may necessitate reconsidering the appropriate application of tags, because funds may revise the narrative discussion in the risk/return summary to better fit the new order.
Given these costs and inefficiencies, the Commission should not require funds to expend the resources necessary to create tagged versions of the current risk/return summary. We recognize that over time additional changes to the risk/return summary may be necessary, and that the taxonomy cannot remain static in perpetuity. In the ten years since the risk/return summary was first adopted, however, only two substantive changes have been made.18 Once the pending changes are incorporated, we have no reason to believe the pace of changes to the risk/return summary will increase going forward. We strongly urge the Commission not to move forward with this proposal while these changes are pending.
B. Necessary Revisions to the Taxonomy Will Take Time
As an alternative to requiring funds to submit XBRL files based on the existing taxonomy, the Commission might consider revising the taxonomy after the Summary Prospectus and ETF Proposals are finalized, perhaps in conjunction with an extension of the compliance date. We do not recommend this approach.
Based on the Institute’s extensive experience drafting the initial risk/return summary taxonomy, we urge the Commission not to underestimate the complexity of updating the taxonomy. Doing so will likely necessitate the addition of tags for new information as well as changes to how existing items are presented.19 The Institute approached the challenge of developing the first taxonomy by assembling a diverse working group of people familiar with fund disclosure, including lawyers, technology specialists and other fund group professionals, software specialists, regulators, consumer representatives, and XBRL experts. Our working group spent numerous hours over the course of approximately eight months considering various methods of tagging data elements. We then made the draft taxonomy available to the public for a 45-day comment period, after which we reviewed and incorporated changes. The taxonomy was further updated in response to comments the Commission received as part of the Voluntary Program comment period, after which it was submitted for acknowledgment by XBRL International. The taxonomy was acknowledged approximately one year after we began the process.20 It has still not received “approval.”
We do not expect that the Commission could revise the existing taxonomy and receive acknowledgement in substantially less time. In fact, because the taxonomy would be proposed for mandatory use, we expect the Commission would receive more—and likely more comprehensive—comments than we received on the existing taxonomy, necessitating additional time for revisions. Nor do we believe the Commission should mandate the use of a taxonomy that has not been acknowledged.
Further, once the taxonomy is completed, funds, software developers and service providers will require time to incorporate the changes into their tagging and viewing tools. Any tools that would adequately assist fund complexes in providing risk/return summaries in XBRL for all of their funds—i.e., any tools with some capacity for automation—would necessarily require a synthesis of the taxonomy into the software. As discussed above, this is not likely to happen until the taxonomy is finalized. Finally, after such tools are developed, funds would need time to explore the available tools, consider the costs, and enter into service agreements before they could begin the process of tagging.
In consideration of these necessary steps, we believe a realistic compliance date could be two years or more away. While the Commission could adopt the current proposal with a flexible or delayed implementation date, we believe the Commission should instead take the opportunity to revise and clarify the proposed rules with respect to funds’ obligations and potential liability, and should consider mandatory tagging of the risk/return summary only after these objectives have been met.
II. The Taxonomy and Tools Have Not Been Sufficiently Evaluated
The Institute does not believe the existing taxonomy and associated tools have been sufficiently vetted for a mandatory environment. The Commission’s observations on this aspect are largely speculative: the Commission “anticipates” that interactive data viewers will become widely available and useful to investors, “believes” software and other technology to detect modifications to the approved tags will be widely available for free or at reasonable costs, and “expects” to develop validation software to check XBRL submissions for compliance.21 Before requiring funds to submit risk/return summaries in XBRL, we urge the Commission to revise and test the taxonomy, and build its validation tool.22 This additional time will also allow improved tagging and viewing tools to be developed. Only then should mandatory use of XBRL for risk/return summary information be considered.
A. The Current Taxonomy Must be Evaluated
The Institute takes great pride in the accomplishments of its XBRL working group, but in creating the first taxonomy to tag large sections of text, we had a novel task before us. Many questions regarding the most effective use of tags, and suggestions for improvements and modifications, have yet to be addressed. In adopting the Voluntary Program, the Commission specifically recognized the existence of such issues, but stated that “commenters did not suggest that the voluntary program should be delayed unless the taxonomy is modified,” and explained that “the purpose of the voluntary program is to test and evaluate tagging technology” (emphasis added).23 Many questions remain open for such testing and evaluation.
For example, the group spent many hours debating whether, for the “strategy” item, the taxonomy should include possible tags for every imaginable strategy, or whether the strategy narrative should just be tagged “strategy.” The former could be far more useful to end users who might wish to search for funds that rely on a certain strategy; the latter would only enable users to view the strategy narratives once they selected funds to compare. On the other hand, generating a comprehensive but not duplicative list of strategy tags proved problematic. Funds often use different words to describe similar strategies, and some working group members expressed dissatisfaction at the notion of being forced to use an externally imposed label. Similar issues arose with respect to the objectives and risk disclosures, as well as with the question of whether we could create certain broad categorizations for funds, much like the “style boxes” of well-known fund information service providers.24
Cognizant that the taxonomy was intended only for testing through voluntary use, the working group ultimately took the simpler approach of single “strategy,” “objective,” and “risk” tags. The group never reached consensus that this was the best or most useful approach, and it was always presumed that these and other important questions would be revisited, ideally with investor input. The Proposing Release does not discuss these fundamental open questions. Moreover, while some testing has begun in the form of voluntary filings, we are not aware of any formal efforts to gather results from voluntary participants or end users of the information, or otherwise to evaluate the efficacy of the working group’s first attempt. The Commission should not mandate the use of a taxonomy before these issues have been thoroughly considered and resolved, to ensure that the taxonomy is well-designed and useful to end users.
B. The Tools Necessary to Comply with a Mandatory Filing Requirement Should be More Developed
The current tools available to create, view, and submit XBRL files are rudimentary at best. We recognize the Commission’s concern that the development of such tools may be slowed without a requirement that funds utilize XBRL. We note, however, that at the time the Proposing Release was issued, the Voluntary Program had only been effective—and thus, such software had only been in any demand—for ten months.25 This does not seem sufficiently long to warrant the Commission’s impatience with the software development process. Complying with the proposed requirements using existing or even moderately improved tools, such as those that could realistically be available in the next year, is simply infeasible for most fund groups, particularly those with multiple funds.26
Virtually all of the Institute members that participated in the Voluntary Program have expressed dissatisfaction with the available tools. One common complaint is that the tagging process is extremely manual. Every tagged exhibit—whether creating a new one or simply updating an existing one—requires a completely new file. Members reported that they could not “copy and paste” tagged data from one file to another, or start with an existing file and add or change only the relevant parts. Similarly, if they detected an error after the file was completed, they could not go back and fix the error, but instead had to start the entire file over again.
Another complicating factor is that existing viewers do not display the tagged files consistently. The SEC’s viewer is not available until after a file has been submitted. Several participants reported that, after reviewing their exhibits using a commercial viewer, they submitted them to the SEC, only to find that they were displayed differently on the SEC viewer, and needed to be recreated and resubmitted.27 One common error stemmed from confusion about how percentages should be entered (i.e., as decimals or integers), resulting in expenses and performance numbers appearing 10 or even 100 times too high. Participants also reported that certain footnotes either were not recognized or did not display in the correct places, so that they appeared unconnected to the associated information. The SEC’s current validation tool did not identify these problems. As a result of these and other glitches, and the need to recreate the entire exhibit every time an error is found, participants in the Voluntary Program widely agree that the current process is not scaleable, and would be unreasonably burdensome for complexes with multiple funds.28
Institute members are far less sanguine than the Commission about the speed with which useful XBRL tools will be developed for funds. They are concerned that software developers are also overly optimistic about their ability to develop and perfect efficient products in the short term. Incorporating products that are still in their beta-testing phase into funds’ workflow, for use with filings that could result in liability under the securities laws, is not an acceptable option. Although the Proposing Release suggests that “the growing number of software applications available to preparers” demonstrates that the XBRL standard and the risk/return summary taxonomy are ready to be mandated,29 its source for this premise is an SEC press release referencing “at least nine software companies whose products will enable public companies to make quarterly and annual financial reports available in interactive data form” (emphasis added).30
It stands to reason that the tools for public company filers are developing rapidly. The voluntary filing program for public companies has been active for four times longer than the fund risk/return program (since March 16, 2005),31 and the participation level in the public company program has been substantially higher.32 Those tools, however, would not be useful to funds for preparing risk/return summary files. To our knowledge, there are substantially fewer fund products in development. Allowing more time for the fund Voluntary Program to grow seems necessary to drive similar software development. Similarly, the Commission’s contemplated validation software, as described in the Proposing Release, must be further developed before Institute members can have confidence that it will aid the filing and error detection process.
Until reliable and efficient tools that would be needed to comply with mandatory XBRL filings are available—or at least reasonably foreseeable within the necessary timeframe—Institute members oppose a mandatory filing requirement.
C. A Testing Period is Necessary
Even once the revised taxonomy and necessary tools are available, experience demonstrates that a testing period is necessary before XBRL filings become mandatory. As described above, a number of questions and problems have cropped up in the Voluntary Program, which were variously attributable to software errors and user confusion. Questions also arose that necessitated clarifying the user’s manual. We expect other such issues to emerge, particularly once the taxonomy is revised, but also as new tools are developed. Funds would benefit from the opportunity to test the final taxonomy, filing instructions, and available products before they are required to use them and face potential liability for their filings.
If the Commission does not allow for a testing period for the new taxonomy, it is critical that there be a phase-in period of at least one year during which time funds would not be subject to liability for inadvertent errors. We believe the liability safeguards under the Voluntary Program, including the cautionary disclosure, effectively offer funds a safe harbor from those inadvertent errors that are likely to occur in the early stages of implementation, while protecting investors from potentially more serious or repeated errors.
III. There is Insufficient Information About the Costs and Benefits of XBRL
The Institute also believes more information is necessary before the Commission can establish that the benefits of requiring XBRL for the risk/return summary are reasonably likely to justify its costs. Certainly there is reason for optimism—the expanding use of XBRL by operating companies for financial reporting globally suggests that this is the wave of the future. But fund risk/return summary disclosure is very different from financial reporting, and much remains to be explored about the benefits—to funds, investors, other market participants, and the Commission—as well as the costs, which will likely be passed along to shareholders.
In considering this proposal, we have attempted to better understand the costs and benefits of XBRL. Our research, like the Commission’s cost/benefit analysis, was necessarily limited by the simple fact that, to date, only 23 out of about 9,500 funds, representing 16 fund sponsors out of approximately 600, have participated in the Voluntary Program. Because these participants were self-selecting, they may not be a representative sample of the industry.33 As a result, it was difficult to draw any definitive conclusions. We therefore believe that the SEC should continue the Voluntary Program until it and other interested parties can better understand both the costs and the benefits of XBRL.34
A. The Benefits of Requiring Tagging of the Risk/Return Summary Need Further Study
In the Proposing Release, the Commission suggests that the benefits of XBRL could include efficiencies for funds in generating regulatory filings and business information processing, as well improved access to risk/return summary information by investors, analysts and others, for free or at reduced cost. While such benefits may exist long term, they have not been demonstrated at this time.
Institute members are skeptical that using XBRL for risk/return summary information will create internal efficiencies that ultimately result in cost savings. Unlike public company financial data, which has multiple uses—including internal and external business reporting, regulatory and tax filings, and investor information—much of a fund’s risk/return information is only used in the prospectus and on the fund company’s website. Because this information is not repackaged and repurposed like financial data, tagging the data cannot be leveraged for other purposes.
With respect to investors, while it is widely agreed that the information contained in the risk/return summary is among the most important information for fund investors,35 very little is known about the extent to which the proposed program would improve their access to such information. As the Commission acknowledges, very few tools are yet available through which investors can access risk/return data in XBRL. Further, as discussed above, the current taxonomy has not been evaluated to determine whether its methodology is the most useful for investors seeking to extract and sort individual data elements. If not properly designed, the mandatory XBRL program could ultimately fail to meet the Commission’s objectives. It would be far easier to perfect the taxonomy before making the program mandatory; relatedly, it is more likely to get a warm reception from the investing public if it is launched in top form. For these reasons, we think it is imperative that the Commission conduct investor research to better understand how and to what extent investors may use XBRL to review risk/return summary information, and whether improvements might be made to the taxonomy to facilitate those benefits.
Finally, the proposal may also benefit the financial community at large in the form of improved data accuracy. As Chairman Cox has explained, one benefit of XBRL is that it would enable users to “draw directly from the source data filed with the SEC. There is no intermediate step of a data provider or service firm mining the data from EDGAR, and loading it into its own database. There’s much less chance for errors, and much greater flexibility in using the data.”36 We agree that XBRL tagging may result in improved accuracy of the information obtained by data aggregators. Many Institute members have confirmed that much of the data used by third party providers is manually inputted, and that they regularly inspect third party data to ensure the information is accurate; if data aggregators used XBRL data feeds, such reviews would be unnecessary, at least with respect to the risk/return information. It is still unclear, however, whether providing risk/return summary information in XBRL will be more cost-efficient for funds than simply reviewing the information collected by data aggregators.
Based on these uncertainties, we believe that more information is needed about the potential benefits of XBRL for risk/return information—as well as the costs—before the Commission can appropriately mandate its use.37
B. The Costs Need Further Study
Notwithstanding the Commission’s statement that its “multi-year experience with [XBRL] has helped us understand the extent to which a mutual fund would incur additional costs,” it is impossible to estimate with any accuracy the true costs of compliance with the proposal.38 As noted above, financial statement reporting and risk/return summaries are entirely different creatures, and the Commission has had less than one year of experience with the latter. The Commission made a reasonable attempt at cost/benefit analysis, but it was based on only six survey responses from Voluntary Program participants. More importantly, as the Commission has acknowledged, very little is known about the tools that may ultimately be developed to help funds comply, such as automated tagging software, or about the cost structure that software and service providers may impose in a mandatory environment.39 These variables, which may be better understood once more useful tools are developed, could substantially impact the cost of XBRL compliance for funds.
Participants in the Voluntary Program who performed their tagging internally typically received tagging software at no cost. Software providers would presumably charge for such tools in a mandatory environment, but none has yet provided a pricing structure. These tagging programs could be extremely expensive for funds. Similarly, external service providers that created XBRL exhibits for participants reportedly did so at reduced or no cost. Those costs, too, are likely to rise substantially, especially if demand for these services is higher than supply, which is likely if XBRL is required at the end of next year as proposed.
Members also have expressed concern about the possibility of unintended consequences relating to service providers. Participants in the Voluntary Program that relied on service providers typically worked with their current vendors, who are already familiar with the funds’ prospectuses.40 Not all fund service providers are likely to offer this service, and even fewer will offer it with good quality or at a reasonable price. Funds will have to balance the efficiency of using a vendor that has already worked with its prospectuses against costs and the quality of XBRL services.
Similarly, funds that currently use small commercial (non-financial) printers for typesetting and printing are concerned that they will have to change service providers in order to access XBRL services. Funds may also have to change their workflow processes or pay for duplicative services in order to obtain XBRL tagging services from vendors. For example, one member that considered, but ultimately did not participate in, the Voluntary Program expressed concerns that funds might have to hire a vendor for services that are currently performed in-house.41
While funds would not likely realize cost savings by tagging and repurposing risk/return data, over time the tagging process may become less costly if funds can adopt software that applies tags to data earlier in the data’s lifecycle. Vendors have begun to offer solutions in this area, and Institute members believe that this capacity will be developed further in information management systems available to them in the future. Moving to such systems is not simply a question of availability, however. Transitioning a fund family’s information management systems and processes is a complex and expensive task. Members report that doing so costs several million dollars up front, and thousands more in ongoing licensing and servicing costs. Complexes typically make such changes approximately every ten years. Moreover, the decision to change systems is typically driven by a variety of business needs, including those related to fund and investor operations, servicing and accounting, marketing, and shareholder communications. Members were skeptical that legal and compliance needs such as XBRL tagging would, by themselves, cause a fund complex to change its information management systems. Thus, fund complexes that have not already budgeted and planned system improvements are likely to be reliant on less efficient back-end tagging tools and services for many years.
C. Incentives Could Encourage Participation in the Voluntary Program and Development of Associated Tools
Increased participation in the Voluntary Program will not only help the Commission to better assess the utility of XBRL, but should also encourage service providers to develop better XBRL-related services. Since the Commission issued the Proposing Release, several Institute members have indicated their intention to submit XBRL exhibits to learn more about the technology. We believe the Commission could do more to promote such activities.
For example, to encourage participation in the original voluntary filing program for financial statements, the Commission offered certain incentives, including expedited review of registration statements and annual reports.42 When it adopted the Voluntary Program for funds, however, the Commission did not provide any similar incentives, despite the Institute’s recommendation that the Commission offer expedited review of fund exemptive applications or expedited review of initial registration statements on Form N-1A or amendments to registration statements to add new funds or series.43 The Commission should reconsider offering these or similar incentives.
The Commission also could take steps to make the tagging process easier and more beneficial for participants. For example, the Commission could offer incentives for, or sponsor the development of, better tagging software. The Commission’s Mutual Fund Viewer could also benefit from substantial improvements, which might help funds and investors understand the utility of XBRL filings. Making the viewer available to funds before they submit their exhibits could also improve the tagging process, by helping funds avoid resubmissions to correct errors. Finally, the Commission should consider instituting a formal process to obtain and address feedback from participants in the Voluntary Program, as well as investors who use the SEC’s viewer. Comments on the SEC’s viewer are particularly important if, as currently proposed, a fund’s compliance with the requirement is partially dependent on a rendering by the SEC’s viewer.
IV. The Proposed Requirements and Liability Provisions Raise Concerns
Several aspects of the Commission’s current proposal raise significant concerns of a different kind. The proposed requirements under which liability may be imposed are confusing and circular, and the liability safeguards, in fact, offer little protection. The proposal also does not address the status of XBRL files that are not rendered (i.e., converted to human readable form), in their entirety, by software provided by the Commission, as well as other scenarios under which errors might arise. On a more basic level, the absence of certain scenarios from the proposal may indicate a need to further consider whether and how the expected uses of XBRL align with the general framework of securities regulation and the fact that funds make continuous offerings of their securities. Finally, the requirement that funds post XBRL files to their websites is unnecessary, and may cause investor confusion.
A. Many of the Proposed Requirements are Unclear
Simply determining a fund’s obligations under the proposed rules is challenging. As we read it, a fund would be required, under a proposed instruction to Form N-1A,44 to submit an Interactive Data File (“IDF”), defined in Rule 11 of Regulation S-T, to the SEC and post the IDF on its website, if any, “in the manner provided by Rule 405” of Regulation S-T, for any registration statement or post-effective amendment that includes or amends information in the risk/return summary. Rule 11 further defines an IDF as a presentation made “in accordance” with Rule 405. Rule 405 requires that an IDF contain the same information, “no more and no less,” as the Related Official Filing (“ROF,” defined in Rule 11 as the ASCII or HTML format part of the official filing that contains the information to which the IDF corresponds), and “comply with the content, format, submission and web site posting requirements of this section.” Failure to comply with the requirement in Form N-1A would result in the suspension of a fund’s ability to rely on Rule 485(b) to file post-effective amendments.45
This structure raises a number of interpretive questions. For example, does submission “in the manner provided by Rule 405” refer to all of the requirements set forth in Rule 405, or only those relating to submission? If the former, a trivial error in compliance with Rule 405 could mean an XBRL submission is not actually an IDF—hence a violation of the Form N-1A requirements, and the automatic suspension of Rule 485(b).
The detailed nature of Rule 405 creates substantial reason for concern about potential penalties, particularly if a minor violation creates a non-compliant registration statement and a suspension of a fund’s use of Rule 485(b). For example, the seemingly innocuous requirement that the IDF contain “no more and no less” than the information contained in the ROF, and the requirement that no data element contained in the corresponding ROF be changed, create the risk that a small omission could impose serious consequences.46 An immaterial omission could escape detection. What would be the import of having a non-compliant registration statement during that time? Similarly, if a fund’s ability to file under Rule 485(b) were automatically suspended because of undetected non-compliance with Rule 405, and the fund subsequently filed an amendment to its registration statement under Rule 485(b), what would be the status of the amended registration statement? Would it be deemed not to have become effective? More importantly, what would be the status of fund shares sold pursuant to the amended registration statement?47
As discussed below, the proposed protections of Rule 406 do not offer satisfactory answers to these and other questions.
B. The Liability Protections Under Proposed Rule 406 are Unclear
1. Interactive Data Files Submitted to the SEC
We appreciate the Commission’s efforts to protect filers from liability for errors that occur despite good faith and reasonable attempts to comply with the proposed requirements. The provisions, however, may not provide the protections intended by the Commission.
Proposed Rule 406 ostensibly contains three provisions offering protection from liability for an IDF. First, under paragraph (c)(1), an IDF submitted to the SEC will be “deemed to comply” with Rule 405 if the filer “makes a good faith and reasonable attempt to comply with Rule 405” and amends a non-compliant filing as soon as reasonably practicable. Second, paragraph (c)(2) states that an IDF “that complies or is deemed to comply with Rule 405 is not subject to liability under any provisions of the Securities Act, Exchange Act, Trust Indenture Act and Investment Company Act or the rules and regulations under those Acts for failure to comply with Rule 405.” Finally, under paragraph (c)(3)(A), an IDF submitted to the Commission is “deemed not filed or part of a registration statement or prospectus for purposes of Sections 11 and 12 of the Securities Act, is deemed not filed for the purposes of Section 18 of the Exchange Act and Section 34(b) of the Investment Company Act, and otherwise is not subject to the liabilities of these sections” (citations omitted).
Rule 11 defines an IDF as a machine-readable file “in accordance with” Rule 405; a filing that is not in accordance is not an IDF. Thus, the provision that an IDF may be “deemed to comply” if the fund makes a “good faith and reasonable” attempt to do so technically may not apply to such a non-IDF file, and therefore offers no protection at all. The other two provisions also only apply to IDFs, which, under the proposed Rule 11 definition, are “in accordance with” Rule 405. The status of a file with a small error is thus entirely open to interpretation, and could be without any liability protection.
Appropriate modifications could be made to these provisions, perhaps to the definition of IDF, such that an imperfect file could still be an IDF, which may or may not be “deemed to comply” with Rule 405. Even so, paragraph (c)(1) is not particularly helpful. Presumably, it was intended to prevent non-compliant registration statements and the suspension of the use of Rule 485(b) for small or inadvertent errors. Unfortunately, its inherently subjective nature offers little comfort against the objective truth of a non-compliant registration statement. The Commission offers no explanation of what constitutes “a good faith and reasonable attempt to comply,” or how or when such a determination would be made.
This framework creates a substantial risk of private litigation.48 For example, a plaintiff might seek to bring a lawsuit under Section 11 or 12 of the Securities Act based on an inaccuracy an XBRL file which may cause it to not meet the definition of an IDF (i.e., it may not be “in accordance with” Rule 405). If the file is not an IDF, it appears not to be entitled to the liability protections unless it is “deemed to comply” with Rule 405 (subject to the modification described above). A fund in such a case would likely seek to dismiss the lawsuit on the basis that, notwithstanding the error, the fund acted in good faith and reasonably, and therefore the file is an IDF that should be “deemed to comply” with Rule 405 and is protected by Rule 406(c)(3). But whether the fund acted reasonably is likely a question of fact that, even if seemingly crystal clear, may not be easily resolved on a motion to dismiss.49
The second provision also raises questions, in that it appears to state the obvious (an IDF that complies with or is deemed to comply with Rule 405 is not subject to liability for failure to comply with Rule 405). The purpose of this provision is unclear. While it appears designed to shield filers from liability, it could be read to imply that failure to comply or be deemed to comply with Rule 405 does subject filers to liability under the securities laws and regulations. Again, given the lack of clarity about what constitutes “deemed to comply” and how and when that might be determined, this implication is troubling.
2. Interactive Data in Viewable Form
To the extent a rendered XBRL file includes all of the information in the risk/return summary, and is rendered by the SEC’s viewing tool, it would meet the definition of Interactive Data in Viewable Form (“IDVF”),50 would apparently be deemed part of a registration statement,51 and would be subject to liability “in the same way and to the same extent as the Related Official Filing.”52 The Commission’s intentions with respect to this structure are unclear.
As noted above, the ROF is defined as the part of the official filing that contains the information to which the IDF corresponds—in this case, the risk/return summary. Read literally, this structure would suggest that an IDVF is subject to the same liability as the risk/return summary. It is not clear, however, what liability that would be. A fund’s risk/return summary may be subject to liability under Section 11 or 12 of the Securities Act, among others, but only in connection with the full prospectus in which it is contained, and the SAI and other information that is typically incorporated by reference into the prospectus.53 It would be inappropriate to isolate the risk/return summary—or IDVF—from the context of the entire registration statement when considering the liability to which it is exposed.
If IDVFs are in fact part of the registration statement, but can be viewed in isolation, they must be able to incorporate by reference the full registration statement, or explicitly be deemed to be accompanied by it. Otherwise, a fund could be exposed to potential liability under Section 12 of the Securities Act for offering to sell a security based on a prospectus containing material omissions. If the Commission intended that, like the risk/return summary, the IDVF would only be considered in the context of the full registration statement for purposes of liability, it must clarify how this incorporation by reference may occur.
The Commission should also consider the potential risks to investors of providing them with only the risk/return summary, without an explicit reference to the additional information that is contained in the registration statement. One option would be to require an IDVF to include a cautionary disclosure with IDVFs to the effect that additional information is incorporated by reference into an IDVF, along with directions on how to obtain that information, much like the legend proposed for the Summary Prospectus. Alternatively, IDVFs could expressly be deemed omitting prospectuses, rather than part of a registration statement, and be accompanied by cautionary language similar to that included in Rule 482 advertisements, which recommends that investors consider certain factors and informs them how to obtain a full prospectus. Or, the Commission could deem an IDVF not an offering document, because by virtue of being submitted in computer-readable code and rendered or converted using the SEC’s software, it is no longer provided directly by the fund.54
In any event, before mandating XBRL, the Commission must consider the appropriate characterization of rendered XBRL files under the securities laws, and create appropriate regulations to ensure that complying with the XBRL requirements does not cause investors to rely on incomplete information, or funds to violate existing laws based on an inability to provide additional information in conjunction with an IDVF.
C. The Liability Protections Must be Broadened
1. XBRL Rendered through Commission Software That is Not an IDVF
Perhaps a bigger question, and one that is not addressed in the Proposing Release, is the status of information converted into viewable text through Commission software that is not identical to the corresponding ROF. Such a file could exist as a result of an error in the Commission’s rendering tool,55 but more importantly, the very nature of XBRL is to enable viewers to obtain select pieces of information. Under the Voluntary Program, an investor can already use the SEC’s viewing tool to generate a report that compares expenses across several funds, but does not include other information about those funds. Although the proposal’s primary purpose is to make the risk/return summary information more useful to investors,56 in part by allowing them to view only selected data elements, the Proposing Release does not discuss the implications of such selective reports.57
We are concerned that the absence of any discussion of such selective reports may create unintended risks for funds. More importantly, we believe this omission reflects a need to further consider whether and how the expected uses of XBRL align with the general framework of securities regulation. Federal securities laws and regulations have historically sought to prevent registrants from providing their investors with information that, viewed in isolation, may be misleading,58 and to ensure that investors receive the key information they need, along with a notice that more information is available upon request.59 By contrast, the proposed rules would require funds to provide information in a format that is intended to allow investors to view and analyze only the pieces of information they desire. Further, the Commission proposes to prohibit the use of a cautionary disclosure because such a disclosure would be “inconsistent with the proposal that interactive data be part of the related registration statement.”60 The absence even of a disclosure advising investors to seek additional information seems fundamentally inconsistent with the Commission’s longstanding approach to disclosure regulation.
At a minimum, the Commission must clarify a fund’s liability where an investor views only a subset of the available data through the SEC’s viewer, including that a fund cannot be liable for a material omission where an investor chooses to view only select information. We also urge the Commission to reconsider the use of cautionary disclosures. More broadly, we think these issues merit a further exploration of the purpose and nature of XBRL, a better understanding of how and through which media interactive data is likely to be accessed, and additional consideration of whether or how to protect investors from being misled if they choose to view only certain pieces of information.
2. XBRL Rendered Through Non-SEC Viewers
The Commission also contemplates that investors will be able to access information converted into viewable text through software other than that provided by the Commission, or downloaded from EDGAR into their own viewers.61 These files are also not IDVFs, because they are not rendered by the SEC’s viewer. The Proposing Release, however, is silent on whether funds could potentially face liability for such files. Once the data is removed from EDGAR by a third party, funds should not be held responsible for its use or display, just as they are not responsible for any information provided to the public by third party data aggregators today. We believe this was the intent of the proposed rules. We request that the Commission state this explicitly.
3. Interactive Data Files on a Fund’s Website
The proposal does not address what liability, if any, attaches to the IDF that would be required to be posted on a fund’s website. The liability provisions in Section 406 specifically govern only those filings that are submitted to the Commission. As discussed below, we strongly oppose requiring funds to place XBRL files on their web sites, but if this requirement is maintained, the Commission must explain how the proposed liability provisions apply to such files.
4. Updates to XBRL Files
Another issue that is not sufficiently considered in the Proposing Release is the impact of requiring (or not requiring) XBRL files to be updated if a fund files a prospectus supplement under Rule 497. The proposed rules would not require such updates to be made.
Funds engage in continuous offerings, and frequently use the Rule 497 process, rather than amending their registration statements, to update information that is included in the risk/return summary. If a rendered file is not updated when a prospectus is supplemented, it may contain different information from the current prospectus, which could potentially lead to liability, to the extent an IDVF could be deemed to include a material misstatement of a fact that has been revised in the registration statement through the Rule 497 process. Even absent liability, if updates were not required, investors using XBRL data could be accessing stale information.62 As we explained in our comment letter on the Summary Prospectus Proposal,63 we are concerned about the potential for investor confusion arising from the availability of different versions of ostensibly the same information.
Requiring funds to update XBRL exhibits in conjunction with Rule 497 filings imposes yet another burden on funds, and additional EDGAR programming would be necessary to permit exhibits to be filed using the Rule 497 process. If the Commission requires that Summary Prospectuses be updated quarterly, requiring parallel updates to XBRL files would create another disincentive for funds to use the Summary Prospectus.64
On balance, we believe that updates to XBRL filings should be required when the content of a fund’s risk/return summary is supplemented through the Rule 497 process. The Commission should ensure that the appropriate updates to EDGAR are in place and tested to accommodate this requirement before adopting such a requirement.
D. Web Posting of XBRL Files Should Not be Required
Institute members strongly oppose the requirement that funds make XBRL files available on their websites because it is unnecessary and could lead to investor confusion. We do not believe that providing such information on a fund’s website will help data aggregators gather XBRL data, because if funds furnish XBRL files to the SEC, all of the files should be available for downloading directly from EDGAR. A data aggregator would be far more likely to collect data from EDGAR than to comb every fund website looking for XBRL files—especially since each fund would likely have different filing and naming conventions for XBRL files on their sites, making automated searches difficult. Moreover, forcing data aggregators to rely on EDGAR would help ensure that they maintained updated information, especially if a fund supplemented its filing mid-year using the Rule 497 process (assuming XBRL files are updated in conjunction with such supplements as we suggest).
In addition, members are concerned that posting these files on the web without a tool to render them may confuse and frustrate investors, as XBRL files are unreadable without a special viewer.65 An investor who tried to open such a file would find unintelligible computer code. While some funds may wish to provide XBRL files along with access to an appropriate viewer (as many do today with PDF files and Adobe Acrobat Reader, which is free), their election to do so will likely be dependent on the availability, at reasonable cost, of a high quality viewer that will not produce rendering errors. Once such software is widely available, a web-posting requirement may be less objectionable.
V. The Commission’s Proposal to Permit Funds to Submit a Tagged Schedule of Portfolio Holdings is also Premature
As with other aspects of this proposal, the Commission’s proposal to permit funds to submit a tagged schedule of portfolio holdings under the Voluntary Program is premature because the taxonomy does not yet exist.66 In addition, separate from the taxonomy, we offer the following two comments on this aspect of the proposal. First, the Commission does not offer any support for its assumption that investors, financial intermediaries and third party information providers will find tagged portfolio holdings information useful. While such analysis may be unnecessary to support a voluntary system, we recommend that the Commission carefully consider the arguments in favor of such information before requiring funds to tag their holdings.67
Second, contrary to the Commission’s assumption, we do not believe that simply permitting funds to submit only portfolio holdings information, without other financial statement information, will encourage participation in the Voluntary Program, absent any incentives to do so. If they were so inclined, funds could seemingly submit only portfolio holdings under the existing Voluntary Program by submitting Form N-Q. In fact, two funds appear to have done so. The Commission should consider whether any incentives might be offered to funds that participate.
* * *
For all of the reasons set forth above, the Commission should not require funds to provide risk/return summary information in XBRL at this time. The Commission should focus its resources on finalizing the Summary Prospectus, which has the potential to benefit shareholders immediately. The Commission should also finalize the ETF proposal, and then turn its attention to revising the risk/return summary taxonomy to correspond to any revisions to Form N-1A. During this time, the Commission should take steps to encourage participation in the Voluntary Program and the development of tools to facilitate XBRL filing. The Commission should return to mandatory tagging of the risk/return summary only after a revised, fully tested and acknowledged taxonomy is in place and the development of tools to facilitate creation and viewing of tagged files is well underway. Finally, any such reproposal must provide clarity with respect to funds’ obligations and potential liability.
The Institute appreciates the opportunity to comment on this proposal. If you have any questions about our comments or would like any additional information, please contact me at 202/326-5815, Frances Stadler at 202/326-5822 or Mara Shreck at 202/326-5923.
/s/ Karrie McMillan
cc: The Honorable Christopher Cox, Chairman
The Honorable Kathleen L. Casey
The Honorable Elisse B. Walter
The Honorable Luis A. Aguilar
Andrew J. Donohue, Director
Susan Nash, Associate Director
Division of Investment Management
1 The Investment Company Institute is the national association of U.S. investment companies, including mutual funds, closed-end funds, exchange-traded funds (ETFs), and unit investment trusts (UITs). ICI seeks to encourage adherence to high ethical standards, promote public understanding, and otherwise advance the interests of funds, their shareholders, directors, and advisers. Members of ICI manage total assets of $12.24 trillion and serve almost 90 million shareholders.
2 See, e.g., Revolution in Real Time: Using the Internet to Inform Investors Better, Remarks at the National Press Club, Paul Schott Stevens, President, Investment Company Institute, Feb. 14, 2006; Securities and Exchange Commission Interactive Data Roundtable, Statement of the Investment Company Institute (June 9, 2006); Letter from Karrie McMillan, General Counsel, Investment Company Institute, to Nancy M. Morris, Secretary, U.S. Securities and Exchange Commission, dated Feb. 28, 2008, available at http://www.sec.gov/comments/s7-28-07/s72807-/pdf//pdf//pdf//pdf/92.pdf; Securities and Exchange Commission Interactive Data Roundtable, Statement of the Investment Company Institute (June 9, 2006).
3 Open-end management investment companies include mutual funds and exchange-traded funds (“ETFs”). Throughout this letter we use the term “funds” to refer to both.
4 See Enhanced Disclosure and New Prospectus Delivery Option for Registered Open-End Management Investment Companies, SEC Release Nos. 33-8861 and IC-28064 (Nov. 21, 2007), 72 Fed. Reg. 67790 (Nov. 30, 2007) (“Summary Prospectus Proposal”).
5 See Letter from Donald J. Boteler, Vice President—Operations and Continuing Education, and Elizabeth R. Krentzman, General Counsel, Investment Company Institute, to Nancy M. Morris, Secretary, U.S. Securities and Exchange Commission, dated March 14, 2007, available at http://www.sec.gov/comments/s7-05-07/s70507-/pdf//pdf//pdf//pdf/6.pdf; Extension of Interactive Data Voluntary Reporting Program on the EDGAR System to Include Mutual Fund Risk/Return Summary Information, SEC Release Nos. 33-8781 and IC-27697 (Feb. 6, 2007), 72 Fed. Reg. 6676 (Feb. 12, 2007) (“Voluntary Program”).
6 See Interactive Data for Mutual Fund Risk/Return Summary, SEC Release Nos. 33-8929, 34-57942, IC-28929 (June 10, 2008), 73 Fed. Reg. 35442 (June 23, 3008) (“Proposing Release”).
7 See Summary Prospectus Proposal; Exchange-Traded Funds, Proposed Rule, SEC Release Nos. 33-8901 and IC-28193 (Mar. 11, 2008), 73 Fed. Reg. 14618 (Mar. 18, 2007) (“ETF Proposal”).
8 See Proposing Release at 53. The Proposing Release indicates that any changes to the risk/return summary—that is, the information currently contained in Items 2 and 3 of Form N-1A—would be incorporated into a revised taxonomy, but suggests that the Commission does not intend to require tagging of all of the information that would appear in the Summary Prospectus. See Proposing Release at 32.
9 See Letter from Donald J. Boteler and Elizabeth R. Krentzman, supra note 5.
10 Extension of Interactive Data Voluntary Reporting Program on the EDGAR System to Include Mutual Fund Risk/Return Summary Information, SEC Release Nos. 33-8823 and IC-27884 (July 11, 2007), 72 Fed. Reg. 39290 (July 17, 2007) at 8. Similarly, Chairman Cox explained that the information gathered from the Voluntary Program would help the Commission “evaluate the proposed interactive formats so we can determine how to implement interactive data most effectively on a broader scale.” Speech by SEC Chairman—Address to the Investment Company Institute’s 2007 General Membership Meeting, May 10, 2007.
11 The Voluntary Program was effective as of August 20, 2007. See id.
12 According to 2008 ICI data, the average fund complex has 21 funds, and 64 complexes have more than 50 funds. Automated processes will have to be developed to enable these complexes to tag all of their funds.
13 In the companion proposal to require public companies to provide financial statement information in XBRL format, the Commission did not propose to include investment companies, at least in part because the taxonomy has not been fully developed. The same principles ought to apply here. See Interactive Data to Improve Financial Reporting, SEC Release Nos. 33-8924, 34-57896 and IC-28293 (May 30, 2008), 73 Fed. Reg. 32794 (June 10, 2008). See also Letter from Donald J. Boteler, Vice President—Operations and Continuing Education, Investment Company Institute, to Florence E. Harmon, Acting Secretary, U.S. Securities and Exchange Commission, dated Aug. 1, 2008 (offering the Institute’s comments on that proposal).
14 Infra Section II.b.
15 Data providers that incorporate XBRL data feeds into their operations would also have to revise their programs to properly download information filed under the revised taxonomy. One major data provider told the Institute that it might not use XBRL data from the current risk/return summary, waiting instead to develop the necessary program until pending changes to Form N-1A are incorporated.
16 Institute members—particularly those who have worked with vendors on voluntary filings—are skeptical that such tools will be available quickly or at reasonable cost. Providers of tagging and viewing software and services are not likely to invest their resources in developing these tools for the current taxonomy knowing that they will become obsolete as soon as a new taxonomy is available. Those that did would presumably pass the development costs on to their clients (i.e., the funds and their shareholders). Funds themselves are similarly reluctant to build automated tools internally, knowing that changes are on the horizon.
17 For example, some risk/return summaries combine the discussion of strategies and risks into a single section. To properly tag such a risk/return summary, one would need to understand how to apply risk and strategy tags to this combined section.
18 In 2001, the SEC adopted amendments to Form N-1A (including the risk/return summary) and related rules to require funds to disclose the effect of taxes on fund performance. See Disclosure of Mutual Fund After-Tax Returns, SEC Release Nos. 33-7941, 34-43857, and IC-24832 (Jan. 19, 2001), 66 Fed. Reg. 9002 (Feb. 5, 2001). In 2006, in conjunction with rules broadening the ability of funds to invest in shares of other funds, the SEC made changes to the fee table in the risk/return summary. See Fund of Funds Investments, SEC Release Nos. 33-8713 and IC-27399 (June 20, 2006), 71 Fed. Reg. 36640 (June 27, 2006).
19 As discussed further below, infra Section II.a., the existing taxonomy has not been refined. At the time it was developed, the Institute and its members believed the Voluntary Program would be in place long enough to assess areas of the taxonomy that need further attention. We do not believe such an assessment has been completed. Doing so may extend the time necessary to complete the revised taxonomy.
20 This process would be required for any change to the taxonomy. To the extent the Commission intends to update the tags from time to time (see Proposing Release at p. 53), we caution that the Commission must plan an extended period for compliance after any rule amendment to allow for the acknowledgment process and then for funds, software developers and service providers to incorporate any changes to the taxonomy into their operations.
21 See Proposing Release at pp. 22, 45, and 43.
22 Our comments are consistent with the Progress Report of the Advisory Committee on Improvements to Financial Reporting, SEC Release Nos. 33-8896 and 34-57331 (Feb. 14, 2008), 73 Fed. Reg. 10898 (Feb. 28, 2008), which stated that, with respect to requiring XBRL for financial statements filed under U.S. GAAP, mandatory implementation may not be possible until, among other things, the U.S. GAAP taxonomy is completed and tested, subjected to public review and comment, and successfully used by voluntary filers for a period of time.
23 Extension of Interactive Data Voluntary Reporting Program on the EDGAR System to Include Mutual Fund Risk/Return Summary Information, supra note 11, at 20-21.
24 While recognizing the value to investors of these externally imposed categorizations, working group members were less comfortable adopting such designations themselves, for use in regulatory filings. For a more detailed discussion of this and other questions considered by the working group, see Remarks at XBRL International Conference, Karrie McMillan, General Counsel, Investment Company Institute, Dec. 7, 2007.
25 The SEC’s Mutual Fund Viewer has been available to the public for even less time—less than four months. See “Mutual Fund Comparisons Using Interactive Data Now Available to Investors,” SEC Press Release, dated April 7, 2008, available at http://www.sec.gov/news/press/2008/2008-55.htm. By contrast, investment companies were able to participate in the Commission’s EDGAR pilot for seven years before the Commission proposed to require filings on EDGAR. See Rulemaking for EDGAR System; Investment Companies and Investment Managers, SEC Release Nos. 33-6945, 34-30952, and IC-18863 (Aug. 7, 1992).
26 Unlike public companies, which only have one set of filings to complete, larger fund families could have a hundred or more funds for which XBRL filings would be necessary.
27 Members also found that, once an exhibit is submitted to the SEC, it cannot be purged from the SEC viewer, even if it has an error. They expressed concern about leaving incorrect files available for public access, particularly because there is no guidance for users on which files to select.
28 See supra note 12.
29 Proposing Release at 21.
30 “SEC’s Office of Interactive Disclosure Urges Public Comment as Interactive Data Moves Closer to Reality for Investors,” SEC Press Release, dated Dec. 5, 2007, available at http://www.sec.gov/news/press/2007/2007-253.htm.
31 See XBRL Voluntary Financial Reporting Program on the EDGAR System, SEC Release Nos. 33-8529, 34-51129, 35-27944, 39-2432, IC-26747 (Feb. 3, 2005), 70 Fed. Reg. 6556 (Feb. 8, 2005).
32 See, e.g., Proposing Release at 6-7 (stating that over 75 companies have participated in the financial statement voluntary program, compared to approximately 20 funds in the risk/return summary program).
33 For example, four of the funds participating in the Voluntary Program (17 percent) were money market funds, although money market funds represent only about nine percent of all funds. These funds tend to have fewer share classes and less complicated risk/return summaries than other types of funds. As a result, the tagging process for them is simpler and faster.
34 A number of Institute members have informed us that they intend to or are in the process of submitting an exhibit under the Voluntary Program.
35 See, e.g., Summary Prospectus Proposal, supra note 7, and Letter from Karrie McMillan, supra note 2, describing a widespread consensus among fund industry participants, investor groups, analysts and others, based on more than a decade of research, including investor research, that the key information about funds that investors want and need includes objectives and strategies, risks, costs, and historical performance.
36 Speech by SEC Chairman: Disclosure from the User’s Perspective—Address to the CFA Institute Conference on Next Generation Asset Management, June 12, 2008.
37 Consistent with our recommendations to the U.S. Treasury Department in connection with its review of the regulatory structure associated with financial institutions (72 Fed. Reg. 58939 (Oct. 17, 2007)), we recommend that even if the Commission demonstrates that the benefits warrant mandating risk/return summary filings in XBRL, it should reexamine the rules after a period of time to “determine whether they are working as intended, whether there are satisfactory alternatives of a less burdensome nature, and whether changes should be made.” Investment Company Institute Submission to the U.S. Department of the Treasury, dated Dec. 7, 2007.
38 Proposing Release at 28.
39 See Proposing Release at 88-89 (“[I]t is unclear how the market for [XBRL] data support services and technology may change if the Commission required over 8,000 mutual funds to submit and post [XBRL files].”).
40 As noted above, significant consideration is necessary to determine how best to apply tags to relevant information contained in a prospectus. Familiarity with the content of a prospectus facilitates this process.
41 For example, vendor-provided XBRL platforms may be driven from a particular desktop publishing or other software program. A fund that prepares its prospectuses for printing in-house and outsources only the actual printing may not prepare the prospectus using the same software as the XBRL vendor. Thus, to use the vendor’s XBRL platform, the fund could have to pay for a conversion service it would not otherwise need.
43 See Extension of Interactive Data Voluntary Reporting Program on the EDGAR System to Include Mutual Fund Risk/Return Summary Information, supra note 11, at n. 77.
44 The proposed instruction, “[a]n Interactive Data File… is required to be submitted…”, creates some ambiguity as to who bears responsibility for the submission. We presume the requirement applies to the fund.
45 The language used in Proposed Rule 485(c)(3) to describe the suspension of a fund’s ability to rely on Rule 485(b) refers to a “registrant.” Many funds are issued as individual series of a single registrant—some registrants may represent 50 or more funds. As drafted, the failure of one fund to comply with the XBRL requirement would result in the inability of its sister funds to rely on Rule 485(b) for post-effective amendments. We believe this approach is too broad and punitive, and would also place form over substance, impacting series funds but not those registered independently.
46 This requirement also presents a curious problem for funds contained in multi-fund prospectuses. The characteristics of such funds may be described in relation to other funds in the same prospectus. For example, one fund complex includes its target date and retirement income funds in the same prospectus. The “objectives” section reads: “The objective of each fund (other than the Income Fund)…”. An investor seeking to view the information about only one fund from this prospectus may be confused by the description of multiple funds. Allowing funds some ability to modify the tagged data for the sake of clarity may improve investors’ understanding of the presented information.
47 For these and other reasons, we question whether suspension of an individual fund’s ability to rely on Rule 485(b) is a necessary or appropriate consequence. The Commission previously considered suspending a fund’s ability to rely on Rule 485(b) if the fund failed to file a report on Form N-SAR for the most recent period for which a filing was required, but did not adopt the proposed limitation after the proposal was criticized as “unnecessary and potentially unfair to funds.” See Post-Effective Amendments to Investment Company Registration Statements, SEC Release Nos. 33-7083 and IC-20486 (Aug. 17, 1994).
48 Similar concerns may also exist with respect to enforcement of the securities laws by the SEC or other government authorities, including under provisions that do not provide a private right of action.
49 For the same reasons, the definition of IDF should avoid using subjective terms such as “materially” in accordance with Rule 405; otherwise the same problem would arise with respect to whether or not a file was an IDF in the first place. For a more detailed discussion of the concerns created when subjective terms are used to describe obligations of a fund that are subject to private rights of action, see the ICI’s comment letter on the Summary Prospectus Proposal, supra note 2.
50 IDVF is defined in Rule 11 as the information displayed when “machine-readable computer code is converted into human-readable text through software the Commission provides; and are … identical in all material respects to the corresponding…[ROF].”
51 Proposing Release at 33-34 (explaining that including a cautionary disclosure (presumably in the viewable format) would be “inconsistent with the proposal that interactive data be part of the related registration statement”). While the rule text does not explicitly state that IDVF is part of the registration statement, Rule 406 specifically excludes only the computer-readable files from being “filed or part of a registration statement” for purposes of certain liability provisions.
52 Proposing Release at 46.
53 Similarly, in the Summary Prospectus Proposal, the Commission proposed to permit a Summary Prospectus to incorporate by reference the prospectus, SAI, and shareholder reports, on the theory that “we do not intend the Summary Prospectus to be a self-contained document, but rather one element in a layered disclosure regime.” Summary Prospectus Proposal at 66.
54 This conversion process is conceptually different from ordinary filings on the EDGAR system, which can be accessed by the public in exactly the same format in which they are filed (i.e., HTML or ASCII).
55 This is not a hypothetical possibility. See supra Section II.b. regarding errors that Institute members encountered with the SEC’s Mutual Fund Viewer during the Voluntary Program.
56 See Proposing Release at 7. We note that this is a somewhat novel approach to the use of XBRL by regulators. The existing uses, as described in the Proposing Release, revolve primarily around the provision of XBRL data for the regulators’ use. See Proposing Release at 14 (referencing XBRL programs by the Federal Deposit Insurance Company, the Office of the Comptroller of the Currency, and the Federal Reserve, as well as foreign countries).
57 Proposed Rule 11 defines IDVF, in part, as human-readable text “displayed… identically in all material respects to the corresponding… risk/return summary information in the [ROF].” Although it is not clear that the Commission intended this result, this language appears to provide that a file containing only selected elements of the risk/return summary is not an IDVF, because it does not contain all of the information in an ROF.
58 See, e.g., Rule 482 under the Securities Act of 1933, limiting the information that may be included in fund advertisements, and requiring certain disclosures, including a statement about the availability of additional information.
59 See, e.g., id; see also Summary Prospectus Proposal.
60 Proposing Release at 33-34.
61 See Proposing Release at 7-8.
62 This would actually be a step backwards from the information investors can obtain today. Data aggregators typically monitor EDGAR daily, and incorporate 497 prospectus supplements into the data they provide. If they relied on XBRL feeds to save time and effort, they would potentially miss such supplements.
63 See ICI’s comment letter on the Summary Prospectus Proposal, supra note 2.
64 See id.
65 We do not read the proposal to require funds to provide access to a viewer for XBRL files. We would oppose such a requirement because it would be costly for funds to license such software and because it would be of limited benefit to investors, since they would only be able to view funds offered by the complex whose website they were viewing. Moreover, many fund websites already provide excellent tools for investors to research and compare funds, often containing data beyond that contained in the risk/return summary.
66 See Proposing Release at 61 (explaining that the Commission “anticipates” entering into a contract to develop a list of tags for portfolio holdings).
67 Although the Proposing Release did not suggest this, the Commission itself may find it useful to receive portfolio holdings information in tagged format. If that is the Commission’s motivation for this aspect of the proposal, the Commission should be forthright in expressing its goals. These benefits, too, should be explored by all interested parties as the Commission considers whether the costs of requiring such filings is warranted.