August 19, 1997
Comments Of Public Citizen On Electronic Records Work Group Draft Report
63 Fed. Reg. 39185 (July 21, 1998)
Public Citizen, Inc., a nonprofit, public interest organization, hereby submits the following comments on the Electronic Records Work Group Draft Report published in the Federal Register on July 21, 1998. The Draft Report was published for comment prior to preparation of a Final Report for submission to the Archivist of the United States.
Public Citizen commends the Work Group for adopting two of the principle conclusions of the Public Citizen v. Carlin decision, namely, that program records must be separately scheduled on an agency-by-agency basis, and that general record schedules for electronic records should specify the time period for disposition. However, Public Citizen believes that the Work Group should not adopt the Draft Report as its final product because the Report's recommendations are fatally flawed in other respects.
Over the past decade NARA has accumulated an abysmal record on electronic records preservation issues, beginning with its insistence in 1989 that agencies' electronic mail communications and word processing records are not "records" under the Federal Records Act.(1) The Archivist should now take action to show that NARA is serious about fulfilling its responsibility to assure that electronic agency records that are valuable to historians, researchers and the public are properly managed and preserved. The recommendations in the Draft Report fall short in this regard. Specifically, the Draft Report fails to give adequate consideration to the Archivist's legal obligation to ensure that valuable records, including those in electronic format, are promptly scheduled, appraised and, where appropriate, preserved. The proposal does not set forth any criteria for identifying records that should be preserved in electronic form and does not give agencies guidance on how to distinguish valuable from invaluable records. Instead, agencies are encouraged to make decisions on destroying or retaining records based solely on their business needs and technical capabilities.
Moreover, the proposal does not give priority to identifying important records that should be preserved in electronic form and assuring that they are appraised and preserved. Instead, the proposal contemplates that agencies may destroy records without obtaining a properly approved disposition schedule, despite the statutory mandate against such destruction. Aside from accepting the district court's decision in Public Citizen v. Carlin that general records schedules be limited to administrative records, and that such schedules should specify a retention period, the Draft Report offers little to move NARA's policies on electronic records forward.
The Archivist should adopt policies and procedures that (i) help agencies identify electronic records that are likely to have sufficient value to warrant long term or permanent preservation; (ii) place a priority on scheduling electronic records that are most likely to have long-term value; (iii) direct agencies on how to implement electronic record keeping for such records; and (iv) provide substantive guidance to agencies on how to prepare schedules for these records. Without such initiatives, the Archivist will continue to be seen as abdicating his responsibilities in this area.
The Work Group's Draft Report adopts two principles for the disposition of electronic records that Public Citizen believes are very important and has worked hard to defend: (1) individual agency disposition schedules, not general schedules, must be used for the disposition of program and unique administrative records; and (2) general records schedules for administrative records that are common to all or several agencies must specify the period of retention. These principles reflect sound archival policy and, moreover, are legally required by the statutes governing disposition of federal records. See Public Citizen v. Carlin, D.D.C. No. 96-2840, opinion (Oct. 22, 1997). General Records Schedule 20 was inconsistent with both of these principles, and it would be a positive, if belated, step forward for the Archivist to implement new records disposition policies that adopt these principles.
Aside from adopting these two principles, however, the Draft Report does little, if anything, to improve NARA's policies or guidance concerning management and preservation of electronic records. Most importantly, the Draft Report does not directly address the Archivist's responsibility to ensure that valuable electronic records are not prematurely destroyed. The Archivist's critical responsibility under the Records Disposition Act is to ensure that the value of the records to the public, researchers, and others is properly evaluated. American Friends Serv. Comm. v. Webster, 720 F.2d 29, 41, 65 (D.C. Cir. 1983); S. Rep. No. 98-373, 98th Cong., 2d Sess. 10 (1984), reprinted in, 1984 U.S.C.C.A.N. 3865 ("Our ability to preserve and make available a full and rich documentary history depends on archivists determining what records are of lasting value and which are not.") The Draft Report, however, gives little, if any, attention to this responsibility and, consequently, does not address the most important issues concerning management and disposition of electronic records.
For example, the Draft Report does not provide any criteria or guidance on how the Archivist should distinguish between electronic records that are disposable and those that have historical or research interest that warrant extended or permanent preservation. To the contrary, the Draft Report focuses exclusively on the "business needs" of agencies and agencies' interests in obtaining disposition authority "[f]rom a Government operations standpoint." 63 Fed. Reg. at 39186, 39188. Consideration of agency's "business needs," however, is not sufficient to permit the Archivist or agencies to determine how electronic records should be scheduled, or what retention periods are appropriate, because appraisal of the records requires consideration of other aspects of the records' value. See NARA, Disposal of Federal Records IV-2 - IV-4(1992); NARA, Disposal of Federal Records 14-15 (1981 ed.).
As a result, the Draft Report provides little to improve NARA's notoriously poor record in addressing electronic record management issues. With respect to the researchers, historians, and other members of the public whose interest the Archivist is charged by statute to protect, the Draft Report provides no standards or criteria to assure that these interests will be properly considered in the future. The widespread perception that the Archivist has abdicated his responsibility to protect these interests, which was underscored by the Public Citizen v. Carlin decision, will only be reinforced if the Archivist adopts the recommendations of the Draft Report in their present form.
With respect to agencies, we respectfully submit that adoption of the Draft Report would do little to improve the conditions that have fostered discontent with NARA's approach to electronic records. The Draft Report expressly disavows any effort to provide agencies with current guidance on electronic record keeping but, instead, recommends that a "follow-on group" be created to formulate such guidance. 63 Fed. Reg. at 39203. For the Archivist to adopt a policy that fails to provide any guidance that agencies can use on this issue now would only perpetuate NARA's failure to provide leadership or meaningful direction in this area.
Moreover, the Draft Report does not provide substantive guidance or criteria to help agencies determine the appropriate retention periods for electronic records. In order to know which records are most important and what types of retention periods are appropriate, agencies must have substantive guidance on how electronic records should be appraised. Because the Draft Report lacks such guidance, agencies preparing the disposition schedules contemplated by the Report will be left to guess how to draft the schedules so that they meet the statutory standard. In short, agencies will be required to submit schedules for NARA's approval or rejection without any meaningful information on what standards NARA will apply. As a result, the management and scheduling of electronic records will continue to be marred by long delays, uncertainty, and confusion.
The scheduling scenario proposed in Appendix C of the Draft Report also does not serve the public or agencies well. To protect the public's interest, it is important to ensure that the electronic records that are most likely to have long-term or permanent value are scheduled promptly and placed under management programs that ensure that they will be preserved. The Draft Report, however, does not encourage agencies to schedule their most important records promptly. Instead, the Draft Report contemplates that agencies should not submit schedules for their unscheduled electronic records until they have completed a comprehensive, series by series review and prepared schedules that cover all of the agencies' records, regardless of the records' value. Because it fails to place a priority on scheduling important records, the Draft Report is inconsistent with the Archivist's responsibility to ensure that agencies promptly identify such records, properly manage them, and take adequate steps to preserve them. Indeed, placing a priority on scheduling such records is particularly important because for many years NARA has discouraged or preempted agencies from submitting schedules for these electronic systems, first by contending that office automation records are not "records" that must be scheduled, and later by adopting GRS 20.
We also think that the Draft Report's proposal will create significant problems for agencies. First, the Draft Report contemplates that some agencies will submit a completed scheduling package to cover all electronic source records within 180 days of NARA's adoption of a Bulletin calling for such schedules. 63 Fed. Reg. at 39190. Agencies adopting this option will be precluded from disposing of any electronic source records for a lengthy, indeterminate period of time. During the period that the agency is preparing this scheduling package, and during the period that the agency is awaiting review and approval or rejection of the package from NARA, the electronic source records will be unscheduled records and, consequently, cannot be disposed of or destroyed. By requiring agencies to submit schedules covering all agency electronic records at once, the Draft Report will impose substantial hardship on agencies that must retain large volumes of electronic records while they await action from NARA.
Second, the Draft Report states that "[i]f an agency cannot determine which model is most appropriate or prepare the necessary submissions within 180 days," the agency may, within 120 days, request "interim disposition authority for a period not to exceed two years for all electronic source records." 63 Fed. Reg. at 39,190. NARA, however, does not have any statutory authority to grant "interim disposition authority." Moreover, allowing agencies to destroy records for two years while they prepare schedules to justify disposal is contrary to the Archivist's responsibility under the statute. The Draft Report's suggestion that agencies may be permitted to destroy records under such "interim authority" is, we believe, misleading, and agencies that rely on this option would be destroying records without complying with the Records Disposal Act.
The criteria that the Draft Report identifies for when agencies may choose to seek interim authority underscore our concern that such requests may lead to the destruction, or continued mismanagement, of valuable electronic records. The proposal indicates that an agency should seek this "interim" authority to destroy records without a complete records schedule if (i) the agency's record schedules are already "out of date" and/or (ii) the agency has not complied with NARA's regulations for preserving electronic mail and other electronic records, most of which were promulgated in 1990. See 63 Fed. Reg. at 39190, column 3. Thus, the proposal contemplates awarding the most leeway to those agencies that have not complied with the law in the past, and have little or nothing in place to safeguard their electronic records against improper destruction.
Specific Comments on Appendix C
In addition to the general comments above, Public Citizen offers the following responses to the specific questions concerning Appendix C, the Proposal for Developing Records Schedules That Include Office Automation Records:
C1. Are the instructions for conducting the analysis of record keeping needs and capabilities sufficiently clear?
No. We do not believe that the instructions are clear and, more importantly, the instructions fail to ensure that the analysis will provide adequate information to appraise the records and determine an appropriate retention period. The Draft Report's instructions on this issue focus exclusively on agencies' business needs, which is only part of the analysis that must be performed to prepare an appropriate records schedule.
C2. Are the instructions for scheduling the records sufficiently clear?
No. The instructions are not clear and, as discussed above, fail to provide substantive guidance to agencies on evaluating the value of records so that agencies can determine (i) how to define series of records so that records with similar value are scheduled together, (ii) what retention periods are appropriate for various types of records, or (iii) what types of records may be appropriate for long-term or permanent preservation.
C3. This document proposes a deadline of 180 days for agencies to submit schedules (SF 115s) to cover their electronic source records or, if the agency cannot meet that deadline, a deadline of 120 days for submitting a plan that sets milestones for accomplishing the scheduling effort. Are these appropriate time frames?
As discussed above, we do not believe that these time periods or the Draft Report's premise that agencies should not submit any schedule until they have prepared a disposition schedule for all unscheduled records are appropriate. As we understand the 180-day deadline, it will have the effect of precluding an agency from disposing of any previously unscheduled electronic records until after the agency prepares its complete schedule with a series-by-series review, NARA appraises the complete schedule, the public comments on the complete schedule, and the Archivist reviews and approves of the complete schedule. We believe that this cumbersome and lengthy process will require the unnecessary retention of many agency records and delay the proper scheduling and management of the most important records. The 120-day period for submitting a plan contemplates that agencies will receive disposition authority more quickly, but it appears to be based on the erroneous premise that NARA can grant "interim disposition authority," even though there is no statutory authority for such disposition authority.
C4. This document includes a questionnaire concerning the systems used by an agency, its technical capabilities for record keeping, and the administrative controls used by the agency. Does the questionnaire ask the right questions to permit NARA to appraise the electronic source records? Should any questions be added or dropped?
As noted above in our general comments and response to Question C2, the questionnaire is inadequate because it fails to focus attention on the value, or lack of value, of the records to those outside the agency. The focus on technical capabilities is also misplaced because the technical capabilities of agency systems are not static. In appraising records and proposed dispositions, the current technical capability is not the only consideration; the potential technical capabilities must also be considered. Moreover, the concepts of a record keeping system and electronic record keeping system are not used consistently in the Draft Report and recommendations, and these concepts are never tied to the considerations underlying appraisal and disposition. Compare 63 Fed. Reg. 39,193, column 3 with 36 C.F.R. § 1234.2 (1998). Record keeping systems may have many different features and capabilities, and it is not necessary for a particular system to have all the features that a record keeping system may have in order for the system to perform valuable record keeping functions.
C5. The scheduling process described in Appendix C will allow NARA to assess proposed retentions for the electronic source records based on the agency's internal records management policy, current record keeping systems and currently installed technology. However, technology changes rapidly and changes in technology will affect record keeping. Currently agencies are required to schedule all "records of new programs and of programs that are reorganized or otherwise changed in a way that results in the creation of new or different records within 1 year of creation" (36 CFR 1228.26(a)(2) and 1234.32(a)). Agencies are also required to review and update their schedules annually (36 CFR 1228.50(d)). Are these requirements appropriate in the current technological environment? If not, what process(es) should be instituted to ensure that as technology changes, agencies and NARA address the issue of whether the retention requirements, and in the case of permanent records, the transfer medium, should be changed?
This question appears to assume that agencies follow NARA's regulation and do, in fact, review and update their schedules annually in accordance with 36 C.F.R. § 1228.50(d). We doubt that this assumption is accurate. Before NARA can evaluate whether an annual review process is adequate for electronic records, NARA must determine whether agencies are following the annual review requirement for records disposition schedules and, if not, what steps must be taken to ensure that agencies perform reviews on the schedule that NARA sets forth in its regulations.
General Comments on Appendix D
With respect to Appendix D, the Proposal to Revise the Entire GRS To Cover All Formats of the Administrative Records Included Therein, we have two specific comments.
First, we disagree with the assumption that publication of the Work Group's appendix satisfies the obligation to publish changes in the General Records Schedules for notice and public comment. The provisions of the General Records Schedules are agency rules and should be promulgated in accordance with the Administrative Procedure Act's procedures for adoption of agency rules. The revisions set forth in Appendix D of the Work Group's Draft Report have not been proposed by NARA or the Archivist or, for that matter, adopted in final form by the Work Group itself. Consequently, the publication of the Work Group's Draft Appendix cannot fulfill the obligation to publish notice of the proposed rule and is more analogous to a notice of inquiry or advance notice of proposed rule making that precedes publication of a proposed rule. The publication of the Appendix also does not provide all of the notice and background information that normally accompanies notices of proposed rulemaking. See Administrative Conference of the United States, A Guide to Federal Agency Rulemaking 171-87 (2d ed. 1991).
Second, while we do not disagree with the Work Group's general idea that the existing general schedules can be modified to provide for the destruction of the electronic versions of many records described in the schedules, we do not believe that adding an instruction that simply states "Destroy/delete after the record keeping copy has been produced," and allows the agency to retain the record "for a limited period of time for operational purposes other than record keeping, such as updating," is an appropriate method to pursue this goal. The instruction lacks specificity and does not appear to comply with the statute's mandate that general schedules provide for disposition after lapse of a specified period of time. 44 U.S.C. § 3303a(d). For example, the proposed instruction permits agencies to retain records for a "limited period of time," but leaves it entirely to the agency to determine whether a limited period is a month, a year, or a decade. Moreover, the instruction is anomalous if the agency's record keeping system stores records in electronic format. If the record has been stored in an electronic record keeping system, it is difficult to understand the justification for the instruction allowing the agency to retain the electronic source records for an unspecified period of time.
In short, we believe that this proposal requires more careful drafting, more specificity, more explanation of the justification for the instructions, and a full opportunity for public comment.
Specific Comments on Appendix E
With respect to Appendix E, the Proposed General Records Schedule for Information Technology Records, we believe that the Work Group's Appendix is generally an improvement on past general schedules for electronic records because it does not include program records, contains specific retention periods, and uses more specific categories than past schedules for such records.
We also agree that Appendix E may not be adopted as a General Records Schedule until after it has been published for comment by the Archivist as a proposed schedule, and the Archivist has considered the comments submitted during any such additional comment period. See 63 Fed. Reg. at 39196.
The Archivist should not adopt the recommendations set forth in the Electronic Records Group Draft Report in their present form.
Michael E. Tankersley
Public Citizen Litigation Group
1600 20th Street, NW
Washington, D.C. 20009
August 19, 1998 Attorney for Public Citizen, Inc.
1. 1 Armstrong v. Bush, 721 F. Supp. 343 (D.D.C. 1989), aff'd in part, rev'd in part, 924 F.2d 282 (D.C. Cir. 1991).