Confidential business information,

[Federal Register: April 21, 2004 (Volume 69, Number 77)]

[Rules and Regulations]

[Page 21409-21425]

From the Federal Register Online via GPO Access [wais.access.gpo.gov]

[DOCID:fr21ap04-8]

DEPARTMENT OF TRANSPORTATION

National Highway Traffic Safety Administration

49 CFR Part 512

[Docket No. NHTSA-02-12150; Notice 3]

RIN 2127-AJ24

Confidential Business Information

AGENCY: National Highway Traffic Safety Administration (NHTSA), DOT.

ACTION: Final rule; response to petitions for reconsideration; correction.

SUMMARY: This document responds to petitions for reconsideration regarding amendments to NHTSA's regulation on Confidential Business Information. These petitions addressed the provisions relating to information submitted to NHTSA pursuant to the early warning reporting regulation. It also corrects a typographic error in the final rule.

DATES: This rule is effective on May 21, 2004. If you wish to submit a petition for reconsideration of this rule, your petition must be received by June 7, 2004.

ADDRESSES: Any further petitions for reconsideration should refer to the docket number and be submitted to: Administrator, Room 5220, National Highway Traffic Safety Administration, 400 Seventh Street, SW., Washington, DC 20590, with a copy to the docket. They may also be submitted to the docket electronically. Documents may be filed electronically by logging onto the Docket Management System Web site at http://dms.dot.gov. Click on ``Help & Information'' or ``Help/Info'' to

obtain instructions for filing the document electronically. You may also visit the Federal E-Rulemaking Portal at http://www.regulations.gov. Follow the online instructions for submitting

comments.

You may call Docket Management at 202-366-9324. The Docket room hours are from 9 a.m. to 5 p.m., Monday through Friday.

FOR FURTHER INFORMATION CONTACT: For questions contact Michael Kido or Lloyd Guerci. They can be reached in the Office of the Chief Counsel at the National Highway Traffic Safety Administration, 400 7th Street SW., Room 5219, Washington, DC 20590, or by telephone at (202) 366-5263.

Anyone is able to search the electronic form of all comments received into any of our dockets by the name of the individual submitting the comment (or signing the comment, if submitted on behalf of an association, business, labor union, etc.). You may review DOT's complete Privacy Act Statement in the Federal Register published on April 11, 2000 (Volume 65, Number 70; Pages 19477-78) or you may visit http://dms.dot.gov.

SUPPLEMENTARY INFORMATION:

Table of Contents

1. Background

   1. The Notice of Proposed Rulemaking

   2. The Final Part 512 Rule

   3. Petitions for Reconsideration II. Consideration of the Issues Raised by the Petitions for Reconsideration

   1. FOIA Exemption 3

   2. NHTSA's Authority to Issue Class Determinations

   3. Scope of Notice

   4. The ``Presumptions'' of Confidentiality in Appendix B

   5. Determinations of Confidentiality in Appendix C

      1. Claims and Notices Regarding Fatalities, Injuries, and Property Damage

      2. VIN and State (or Foreign Country) Information

      3. Common Green Tires

      4. Production, Warranty Claims and Adjustments, Field Reports and Consumer Complaints

      1. Production Numbers

      2. Warranty Claims and Adjustments

      3. Field Reports

      4. Consumer Complaints

   6. Data Quality Act III. Regulatory Analyses and Notices

1. Background

   1. The Notice of Proposed Rulemaking

      On April 30, 2002, NHTSA published a Notice of Proposed Rulemaking (``NPRM'') to amend 49 CFR Part 512, Confidential Business Information (``Part 512'' or ``CBI''). 67 FR 21198 (April 30, 2002). The agency sought to simplify and update the regulation to reflect developments in the law. The NPRM also asked for comments on whether to

      [[Page 21410]]

      create class determinations covering portions of the data to be submitted under the early warning reporting (``EWR'') rule, (see Subpart C of 49 CFR Part 579), which NHTSA had proposed pursuant to the Transportation Recall Enhancement, Accountability and Documentation (``TREAD'') Act. The comment period closed on July 1, 2002.

      The agency received timely comments from various sectors of the automotive industry, including vehicle manufacturers, tire manufacturers, supplier and equipment manufacturers, and other interested parties. Comments were received from the following trade associations: The Alliance of Automobile Manufacturers, the Association of International Automobile Manufacturers, the Rubber Manufacturers Association, the Tire Industry Association, the Motor and Equipment Manufacturers Association and the Original Equipment Suppliers Association, the Automotive Occupant Restraints Council, the Juvenile Products Manufacturers Association, the Truck Manufacturers Association and the Motorcycle Industry Council. Comments were received also from individual manufacturers: General Motors North America, Cooper Tire, Utilimaster, Blue Bird Body Company, Bendix, Harley-Davidson Motor Company, Hella North America, WABCO North America, Meritor-WABCO, and Workhorse Custom Chassis. Enterprise Rent-A-Car Company and the Washington Legal Foundation also filed comments. Individual requests for confidential treatment for all EWR submissions were also received by several trailer manufacturers and from the Truck Trailer Manufacturers Association. Public Citizen also filed comments on November 27, 2002, as well as supplemental comments thereafter. The agency considered all comments when promulgating the final CBI rule.

      The vast majority of the comments addressed whether the various categories of EWR information should be treated confidentially. Public Citizen argued that all information should be disclosed. Business interests argued that some or all of the data should be withheld from disclosure, claiming either that Congress intended for the agency to withhold all early warning reporting information or that disclosure would cause substantial competitive harm or result in less information being available for the agency's early warning detection program.

   2. The Final Part 512 Rule

      The final CBI rule specifically addressed the disclosure or confidentiality of EWR data.\1\ 68 FR 44209, 44216 et seq. (July 28, 2003). The agency determined that the TREAD Act's provision on disclosure of EWR information (49 U.S.C. 30166(m)(4)(c)) did not allow withholding all EWR information from disclosure under Exemption 3 of the Freedom of Information Act (``FOIA''), which incorporates nondisclosure provisions contained in other federal statutes. The agency concluded that Section 30166(m)(4)(c) was not intended to foreclose the application of FOIA Exemption 4 to determine whether certain data should be disclosed under FOIA, but rather was intended to make more stringent the showing necessary for the agency to disclose otherwise confidential information.

      \1\ The agency also set out the procedures to follow in seeking confidential treatment for information generally. Section 512.21(c) of those procedures contained a typographical error. After referring to the Chief Counsel's denial of a petition for reconsideration of the denial of a request for confidentiality, the rule states that ``the information may make the information available.'' We are correcting this to state that once a petition for reconsideration under Part 512 has been denied, ``the agency'' may make the information publicly available.

      The agency determined that some, but not all, of the types of information required under the EWR rule would be withheld from disclosure pursuant to Exemption 4. This exemption is applicable to certain confidential business information that, if disclosed, would likely cause substantial competitive harm, impair the government's ability to obtain information in the future, or both. The agency created class determinations applicable to EWR information pertaining to production numbers (except those for light vehicles), warranty claims, field reports, and consumer complaints. Those class determinations were added as Appendix C to 49 CFR Part 512. Further, the agency concluded that the disclosure of certain categories of EWR data is likely neither to cause substantial competitive harm nor to impair the government's early warning detection program. Accordingly, the agency decided against creating class determinations covering EWR information relating to fatality and injury claims and notices and to property damage claims.

      The agency retained the class determinations on confidentiality in Appendix B, which have been applied for years to blueprints and engineering drawings containing process of production data (under limited conditions), future specific model plans (until the first model is offered for sale), and future vehicle production or sales figures for specific models (until the applicable model year production period ends). The agency revised the language of Appendix B to provide that such materials are determined entitled to protection under FOIA Exemption 4, as opposed to the historical language providing that such materials were presumed to be entitled to such protection.

   3. Petitions for Reconsideration

      The agency received three timely Petitions for Reconsideration, one each from the Rubber Manufacturers Association (``RMA''), the Alliance of Automobile Manufacturers (``the Alliance''), and Public Citizen Litigation Group on behalf of the Trauma Foundation, the Consumer Federation of America, Advocates for Highway and Auto Safety and the Center for Auto Safety (``PCLG'').

      The RMA asks the agency to hold all of the EWR information confidential. It reiterates its position that Section 30166(m)(4)(c) qualified as a FOIA Exemption 3 statute prohibiting the release of any EWR information submitted to the agency and argues further that the release of this information would violate the Data Quality Act, 44 U.S.C. 3516. The RMA also makes further arguments in support of its position that fatality, injury and property damage claim information should be accorded class treatment under FOIA Exemption 4 and sought clarification of the agency's intended treatment of EWR reports relating to common green tires.

      PCLG, on the other hand, asks the agency to vacate all the EWR class determinations in Appendix C and to release all of the EWR information to the public. PCLG reiterates the view expressed previously by Public Citizen in its comments on the NPRM that the purposes of the TREAD Act can only be achieved if all of the EWR information is available to the public.\2\ PCLG claims the NPRM did not provide sufficient notice that the agency would consider the creation of class determinations or change the language of the pre-existing class determinations. PCLG also argues that the agency lacks authority to create class determinations, and further that

      [[Page 21411]]

      individualized showings are necessary before any data are withheld.

      \2\ In the Final Rule, we explained that the agency and Public Citizen differed in their views of the purposes of the TREAD Act. Public Citizen, and now PCLG on behalf of the petitioning organizations, contends that the early warning provisions of the TREAD Act were intended to supply the public with vast amounts of information collected from manufacturers. NHTSA believes that the provisions were intended to enhance the information available to the agency from which it could promptly identify potential problems.

      The Alliance requests that the agency reconsider its anticipated treatment of vehicle identification numbers (``VIN'') in EWR reports relating to fatalities and injuries. According to the Alliance, information is readily available over the Internet from which personal identifiers can be discerned using the complete VIN information. On this basis, the Alliance requests that the agency withhold from public disclosure complete VIN information pursuant to FOIA Exemption 6. The Alliance also asks that the agency withhold from public disclosure information relating to the state in which a reportable incident occurred as well as information on the country if the incident relates to an event that occurred outside of the United States. Again, the Alliance claims that state and foreign country information, when combined with other data, can lead to the revelation of personal information.

2. Consideration of the Issues Raised by the Petitions for Reconsideration

   1. FOIA Exemption 3

      The RMA reasserts its comment on the NPRM that Section 30166(m)(4)(c) is a statutory prohibition against the disclosure of any early warning data unless and until a defect or noncompliance investigation has been opened by NHTSA. RMA adds no new information to support its position.

      As set forth in the detailed analysis accompanying the final Part 512 rule, the agency has concluded that Section 30166(m)(4)(c) does not qualify as an Exemption 3 provision. The case law makes clear that to satisfy Exemption 3, a law must either require that matters be withheld from the public in such a manner as to leave no discretion on the issue or establish particular criteria for withholding information or refer to types of matters to be withheld. In either instance, the level of discretion afforded to the agency must be severely restricted, a situation that is not evidenced by the language of 49 U.S.C. 30166(m)(4)(c). This statutory provision instructs the Secretary to determine initially which of the early warning reporting information is entitled to confidential treatment as confidential business information and, if so, then to consider whether disclosure will assist in the agency's implementation of the defect and remedy provisions of the Act. See 68 FR at 44225-44226. Among other things, the Secretary's decision whether the disclosure of the information will assist in carrying out those other statutory provisions is highly discretionary.

   2. NHTSA's Authority To Issue Class Determinations

      PCLG asks the agency to reconsider its use of the ``class determination'' device, arguing that the agency lacks the authority to issue regulations treating like information as categorically subject to a FOIA exemption. According to PCLG, where Congress wants to exempt a category of records without requiring submitters to satisfy FOIA Exemption 4, it has exempted the information by statute.

      PCLG argues that the agency may not treat any submission as subject to a FOIA exemption unless the submitter has made an individual showing that disclosure of the particular data meets the requirements of FOIA Exemption 4. Under its approach, the agency would have to review each EWR submission from each manufacturer regarding each reported item of data for each reporting period individually.

      The agency disagrees. PCLG would require individual reviews despite the long history of class determinations, and the facts that numerous EWR reports containing the same informational elements for each category of manufacturer under 49 CFR 579.21-579.26 are submitted pursuant to standardized electronic reporting templates and that the information elements do not change from reporting period to reporting period. Because each data submission contains the same elements of information, in the same format (as required by the regulation), decisions relating to the disclosure of the data will not vary. As a result, individualized determinations will merely impose an administrative burden on the agency and manufacturers that can be avoided through a class determination.

      The agency first proposed class determinations in a 1978 NPRM and adopted them in a final rule issued in 1981. See 46 FR 2049 (Jan. 8, 1981). During this early rulemaking, NHTSA made clear that a key purpose of the class determination was to improve its efficiency in processing requests for confidential treatment with regard to sufficiently specific categories of information:

      Although making class determinations relating to business confidential information is a difficult undertaking (as evidenced by the fact that few agencies make such determinations), the agency believes that to the extent that such meaningful classes can be identified and described, class determinations will ease the burdens of both the agency and submitter of information in making and processing claims for confidential treatment of information.

      43 FR 22412, 22414 (May 25, 1978).\3\ NHTSA also stated that the process would benefit the public by making information not subject to a FOIA exemption available more quickly. 46 FR 2049.

      \3\ The Freedom of Information Clearinghouse (``Clearinghouse''), a joint project between Public Citizen and the Center for Responsive Law, commented on the original CBI rule, noting that the group generally supported the proposed rulemaking but expressed reservations over the application of class determinations unless the determinations were rebuttable and did not act to limit the authority of the Administrator to release that information under limited conditions. Comments from the Freedom of Information Clearinghouse, Docket 78-10; Notice 1, No. 10, at 3 (July 28, 1978). NHTSA incorporated these suggestions into the final rule. The Clearinghouse raised similar concerns during a subsequent Part 512 rulemaking that addressed, among other things, the confidentiality of cost data as a class. Comments from the Freedom of Information Clearinghouse, Docket 78-10; Notice 9, No. 5, at 5 (Aug. 21, 1989). Provisions allowing the Administrator to make otherwise confidential information public remain today, and the disclosure provision in the TREAD Act addresses that process.

      Thus, since 1981, NHTSA's regulations have included a provision (49 CFR 512.10 (2002)) declaring the authority of the Chief Counsel to issue class determinations. Consistent with this authority, Appendix B to Part 512 has long included three class determinations that identify certain classes of information as presumptively resulting in substantial competitive harm to a submitter if disclosed.

      Class determinations of confidentiality are not unique to NHTSA. Class determinations contained within Food and Drug Administration regulations cover certain information that the agency receives. See, e.g. 21 CFR 20.111(d). Similarly, the Environmental Protection Agency has established through regulation a process through which it creates class determinations, 40 CFR 2.207, and has created a number of class determinations that cover specified information. Like the EWR data received by NHTSA, the information covered by these regulatory regimes is not generally subject to a statutory prohibition on disclosure that satisfies FOIA Exemption 3.

      In their interpretations of FOIA, courts have encouraged the development of categorical rules whenever a particular set of facts will lead to a generally predictable application of FOIA. See, Critical Mass Energy Project v. NRC, 975 F.2d 871, 879 (D.C. Cir. 1992) (en banc). In Critical Mass, the court noted that establishing a discrete category of exempt information will implement the

      [[Page 21412]]

      congressional intent to provide workable rules and that such categorical rules further FOIA's purpose of expediting disclosure. Id.

      Courts have not questioned the authority of agencies to promulgate regulations involving confidentiality under FOIA. See Neal-Cooper Grain Company v. Kissinger, 385 F. Supp. 769 (D.D.C. 1974) (discussing agency regulation that protected certain categories of information from disclosure), and EEOC v. Associated Dry Goods, 449 U.S. 590 (1981) (upholding the validity of an agency's regulation that permitted limited disclosures of case information to the relevant parties, their attorneys, and witnesses as necessary for the agency to carry out its functions under Title VII of the Civil Rights Act).\4\

      \4\ See also, O'Reilly, Fed. Info. Disclosure 3d, section 10.10 (2000) (agencies which have frequent submissions of confidential business data may pre-designate specific classes as confidential.)

      In adopting the final CBI rule, NHTSA made a decision to proceed by rule rather than individual determinations. The choice whether to employ rulemaking or individual adjudications to resolve an issue is one left primarily to the agency, SEC v. Chenery, 332 U.S. 194 (1947), and courts have consistently and favorably recognized the ability of agencies to promulgate regulations without having to resort to individual resolutions or orders. See National Petroleum Refiners Ass'n v. FTC, 482 F.2d 672 (D.C. Cir. 1973).

      There are many valid reasons for proceeding by rule. To begin, the EWR regulation requires the submission of standardized reports, which are particularly well suited to the resolution of confidentiality claims by rule. Its provisions apply to manufacturers of certain types of motor vehicles and motor vehicle equipment. For each type of vehicle or equipment, the reporting elements are identical for all covered manufacturers. The reporting is performed utilizing standardized reporting templates. While the data reflect the individual experience of each manufacturer, the nature of the information reported is the same.

      Whether particular information is entitled to confidential treatment under Exemption 4 depends on the nature of the information and the likely consequences of its release. EWR information from various manufacturers of motor vehicles, child restraints and tires (e.g., the number of warranty claims) should be treated the same way under the law, both because the impact of disclosure on the competitive environment is the same as applied to those manufacturers and because the possibility that releases of the information could lead to more restrictive policies applicable to warranties, field reports and customer complaints is the same. Proceeding by rule achieves consistent resolutions of confidentiality based on criteria under Exemption 4.

      Second, mandating individual requests for confidential treatment would, taking into account in-house experience and capabilities, subject smaller businesses to a disadvantage. We expect that larger manufacturers would routinely seek confidential treatment for EWR submissions, but that many smaller manufacturers (who are less familiar with regulatory practice) would have difficulty in properly assembling and submitting the material that must accompany an individual request for confidential treatment under Part 512. As a result, it is likely that the data submitted by larger manufacturers would be accorded confidential treatment under Exemption 4, but that the same type of data submitted by relatively small businesses would not. The small business would then face the costs of obtaining outside support for an appeal under 49 CFR 512.9. These burdens and costs run against the grain of federal laws and executive orders that seek to reduce, as opposed to increase, the regulatory costs on small businesses. See e.g., 5 U.S.C. 601 note. While we anticipate that, over time, smaller businesses will properly seek such treatment for their submissions and learn how to present a valid claim, in the interim, a small business would be unduly disadvantaged despite the fact that its submissions should be entitled to the same treatment as those of larger and more sophisticated manufacturers.

      Third, the courts have long recognized that agencies have the ability to promulgate those regulations that are necessary for them to perform those tasks Congress has assigned them. See Federal Power Comm'n v. Texaco, 377 U.S. 33 (1964) and In re Permian Basin Area Rate Cases, 390 U.S. 747, 780 (1968). See also Weinberger v. Bentex Pharmaceuticals, 412 U.S. 645, 653 (1973); Balelo v. Baldrige, 724 F.2d 753, 760 (9th Cir. 1984). In establishing the early warning reporting program, Congress directed NHTSA to collect information from manufacturers to assist the agency in promptly identifying possible safety-related defects. 49 U.S.C. 30166(m). Congress recognized that much of the information would be statistical in nature, reporting of the information would be in electronic form, and computer database systems would be used to review and utilize the information.\5\

      \5\ See 49 U.S.C. 30166(m)(3)(A) (providing for the Secretary to collect warranty and claims data, including aggregate statistical data on property damage from alleged defects) and 49 U.S.C. 30166(m)(4)(A) (providing that the Secretary shall specify the form of reporting EWR data, including by ``electronic form''). Congress also told the agency to identify the systems it would employ to review and utilize the information and to take into account the agency's ability to use the information in a meaningful manner to assist in the identification of safety related defects. 49 U.S.C. 30166(m)(4)(A)(ii) and (D).

      Consistent with the statute and to achieve the results Congress expects--the earlier identification of potential safety issues--in a manner that does not require unavailable staffing, NHTSA has required manufacturers to submit large volumes of data in a consistent format that a computer can manage and sort through using statistical analyses. See 67 FR 45865-66. The resolution of the confidentiality of EWR data by class determination rather than by individualized assessment is consistent with this approach. If individualized review of confidentiality requests were required, the limited capacity to review a large number of individual confidentiality requests,\6\ rather than the ability to efficiently assess large volumes of early warning data by computer and make follow-up inquiries, would strongly and negatively influence the scope of the early warning data collection effort. If NHTSA were to tailor early warning reporting to its capacity to manually process confidentiality requests made by individual written requests as opposed to class determination by rule, the program would be constricted and the results contemplated by statute would be compromised.

      \6\ In the first reporting period, early warning reports were submitted by over 50 light vehicle manufacturers, over 70 bus and medium-heavy vehicle manufacturers, over 150 trailer manufacturers, 13 motorcycle manufacturers, 18 tire manufacturers, and 8 child restraint manufacturers.

      Finally, we are concerned that requiring individual requests for confidential treatment would have an adverse effect on the public's ability to access the public portions of the EWR data expeditiously. Were we to require individual confidentiality claims, we expect that manufacturers would make claims for confidentiality covering various EWR submissions. We would need to review and analyze each claim separately. Under Part 512, information that is the subject of a confidentiality claim is withheld from disclosure to the public while the agency considers the claim. The result is likely to be a substantial ``back-log'' of EWR

      [[Page 21413]]

      information in the confidentiality review process. Data that ultimately will be determined to be public will not be identified and made public until the process is complete, on a claim-by-claim basis. Moreover, the data would not be public until manually transferred to the public portion of the agency's data system on an individualized basis, which would entail further delay. The diversion of effort to review confidentiality claims for EWR information would also slow the processing of confidentiality requests covering other sorts of information submitted to the agency and, similarly, it would delay determinations that some of that information is not public and the release of that information to the public.

      This stands in contrast to the system being implemented based on class determinations, which enables the agency to transfer appropriate data directly to the public section of the EWR database promptly following its receipt. In short, consistent with the use of a computerized database contemplated by the TREAD Act, class determinations allow the agency to establish database protocols that automatically protect confidential data while allowing prompt access to non-confidential information. In contrast, a system requiring individual review of every confidentiality claim is likely to delay the public's access to information not protected by a FOIA exemption.

      In sum, we believe that the agency has the authority to establish class determinations categorically covering similar information (as it has done for decades), and that the early warning reporting information (with its standardized reports) is particularly well suited to class determinations. Individual consideration of each early warning submission is not only infeasible, but also would seriously overwhelm agency resources.

   3. Scope of Notice

      PCLG asserts that the agency did not provide adequate notice that it might apply class determinations to EWR data. It asserts the NPRM did not propose the categorical exemptions for EWR information or identify them as an option that the agency was considering, but rather expressed the intent not to add class determinations and to create a presumption of disclosure. Accordingly, PCLG claims the class determinations should be vacated.

      We disagree. The NPRM provided sufficient notice under the Administrative Procedure Act (``APA'') of the agency's considerations with regard to the confidentiality of the EWR information. The APA is intended to ensure that the public has a meaningful opportunity to comment on potential agency action. The case law construing the APA makes clear that a final rule may differ from the proposed rule and that additional information received during the notice and comment period will play a role in shaping the terms of the final regulation.\7\

      \7\ The question whether the initial notice is adequate sometimes is cast in terms of whether a second round of comment is necessary. The test for deciding whether a second round of comment is required is whether the final rule promulgated by an agency is a logical outgrowth of the proposed rule. American Water Works Ass'n v. EPA, 40 F.3d 1266, 1274 (D.C. Cir. 1994). That standard is applied functionally by asking whether the purposes of notice and comment have been adequately served--that is, whether a new round of notice and comment would provide the first opportunity for interested parties to offer comments that could persuade the agency to modify its rule. Id. See also Environmental Defense Center v. EPA, 344 F.3d 832, 851 (9th Cir. 2003) (restating logical outgrowth test).

      The NPRM expressly sought comment on whether to create class determinations with regard to the EWR data, while recognizing that the final EWR requirements had been proposed but not yet been promulgated. After discussing the possibility of creating class determinations applicable to information submitted in response to particular investigations,\8\ the agency sought comments with regard to the treatment of EWR information:

      \8\ Many manufacturers pointed out the distinction between the comprehensive nature of the EWR data and submissions in response to information requests in individual defect investigations for which they generally do not seek confidential treatment.

      We are also interested in receiving comments regarding whether any of the proposed class determinations should be applicable to the material to be submitted under the agency's ``early warning'' regulations and whether any additional class determinations should be established. For example, the agency's ``early warning'' NPRM proposes that manufacturers submit to the agency reports on incidents involving deaths or injuries and copies of field reports. The agency seeks comments regarding whether the agency should presumptively determine that these (or a subset of these) types of documents would or would not cause competitive harm to the submitter if released. Any suggested changes or additions to the proposed list of class determinations should be justified. We recognize that a final rule has not yet been issued regarding the ``early warning'' requirements, but we ask commenters to provide as much information as possible within this comment period. If necessary, we will allow for additional comments prior to finalizing any class determinations

      covering the ``early warning'' submissions.

      There can be no doubt that the public understood the potential scope of the rulemaking. We received numerous comments from a myriad of sources. The comments--including those of Public Citizen--addressed all parts of the EWR requirements and addressed in detail each category of EWR data as a class. This included both whether the various categories of EWR information should be accorded confidentiality and the nature of the determination.

      Public Citizen's comments expressly addressed the possibility of creating additional class determinations, favoring those that would find data presumptively public and opposing those that would find information presumptively confidential. Public Citizen argued that no showings of substantial competitive harm were significant enough to justify the use of class determinations for any EWR information. This supports our conclusion that Public Citizen, as well as other interested members of the public, had adequate notice about the possible application of class determinations to EWR information.

   4. The ``Presumptions'' of Confidentiality in Appendix B

      In the final CBI Rule, the agency amended the preexisting class determinations--contained in Appendix B to Part 512--from determinations that information covered by those class determinations would be treated as presumptively confidential to determinations that the information is protected by FOIA Exemption 4. We did not change the scope of Appendix B, which applies to certain categories of information--blueprints and engineering drawings that contain process and production data, future specific model plans (until the vehicle model is offered for sale), and future vehicle production or sales figures (under limited circumstances) for specific models. PCLG objected to the amendment.

      Upon reconsideration, we agree with PCLG that there is merit to the application of a presumption, as opposed to a determination, for the class determinations in Appendix B. Appendix B is typically invoked by a company in connection with the submission of specific information in response to an agency information request. The agency reviews the materials in light of the claim that the particular information falls within the category of information covered by Appendix B. The submitter also provides the certification required by 49 CFR 512.4, which requires the company to attest that it has in fact maintained the confidentiality of the material at issue.

      [[Page 21414]]

      While not common, it is possible that information a manufacturer claims to be covered by a class determination is not in fact covered by that class determination. For example, a sketch of a component may be claimed to be an ``engineering drawing,'' but in fact may not be specific enough to enable another company to manufacture it. The dimensions and specifications of some commonly used automotive components (i.e. wheel studs, brake linings, suspension components, etc.) may already be in the public domain. A future product plan may have been announced, or may be announced between the time the claim is made and when we review the claim.

      The presumption, coupled with the requirement for an individualized claim, strikes a balance consistent with the possibility that materials submitted may be outside the scope of Appendix B, and the possibility that materials may be in the public domain.

      In light of their application, we have decided that the class determinations in Appendix B had properly provided that the agency has determined that disclosure of data within those categories ``presumptively'' will cause substantial competitive harm. Appendix B is being revised to read as it had before the final rule became effective.\9\

      \9\ In contrast to Appendix B, we continue to believe it appropriate that the class determinations in Appendix C (applicable to EWR data) include a determination that the covered classes are exempt from disclosure, rather than a determination that they are ``presumptively'' exempt from disclosure. Unlike particular submissions responding to specific questions in individual investigations, the EWR data will provide identical elements of information pursuant to the EWR regulation, electronically, at regular intervals. The EWR data do not give rise to the same concern leading us to reestablish the ``presumption'' applicable to other class determinations. There is no issue as to whether it falls within or outside of the category of information covered by the class determination.

   5. Determinations of Confidentiality in Appendix C

      The RMA and PCLG petitions ask the agency to change its treatment of EWR data, taking diametrically opposed positions. The RMA argues that Appendix C to Part 512 should be expanded to include all EWR data not already included in Appendix C, i.e., information about claims and notices of fatalities and injuries, the number of property damage claims, and information about common green tires. In contrast, PCLG asks the agency to vacate all of Appendix C, asserting that the class determinations for certain production data, and information relating to warranty data, field reports and consumer complaints were not justified.

      In addition, the Alliance petition seeks an expansion of the Appendix C class determinations of confidentiality to include two data elements in reports on incidents involving fatalities and injuries-- vehicle identification numbers (``VINs'') and state or country of incident (if outside of the United States). 1. Claims and Notices Regarding Fatalities, Injuries, and Property Damage

      In the final CBI rule, the agency concluded that the information about claims and notices of fatalities and injuries and the number of property damage claims (``claims information'') are not entitled to confidential treatment. We noted that information about such claims is often publicly available, either from court documents or from media reports about crashes. As we explained, this information is not likely to reveal business strategies or other data that can be used competitively. We also found there to be no likelihood that disclosing this information would impair the agency's defect investigation program.

      The RMA petitions the agency to reconsider its treatment of these claims data, asserting that information about fatality, injury or property damage claims is similar in nature to that about warranty claims, field reports and consumer complaints, which are included in the Appendix C class determinations of confidentiality. The RMA also argues that the claims information amounts to unverified or unsubstantiated allegations, preliminary to the determination of a defect, and will be wrongly perceived by consumers and others. It contends that the data may be used in misleading cross-company comparisons, potentially affecting purchasing decisions by consumers, and that this could result in competitive harm. The RMA further asserts that the compilation of information about claims provides a more robust database than might otherwise be publicly available.

      We have considered the RMA's petition, but continue to believe that early warning reporting information on fatality, personal injury and property damage claims does not fall within the purview of FOIA Exemption 4. Unlike the comprehensive disclosure of warranty, field report and consumer complaint information, release of EWR claims information will not reveal underlying business decisions, approaches and strategies. As explained in the final rule, the warranty, field report and complaint information reflect the business policies, practices and decisions (and, in some circumstances, cost structures) of each manufacturer. Disclosure of the comprehensive database of this information would provide competitors with information about how consumers view their products and corporate marketing efforts. They reflect what customers say, like or dislike and seek to have repaired, changed or replaced, providing considerable feedback, by system and component, on product performance and developmental issues.

      In contrast, disclosure of information on fatality, injury and property damage claims does not reveal corporate strategies or intangibles such as consumer acceptance of product features or reaction to corporate programs, such as broader warranty coverage.\10\ The claims data are far fewer in number. They reflect actual events (although the cause and nature of the event and the responsibility for any consequential injury or damage is often disputed) that are historical and do not reflect ongoing and typical customer experiences or product evaluations.\11\ The remainder of RMA's petition appears to be premised primarily on two erroneous beliefs. First, the RMA seems to assume that early warning data will be treated as evidence of a safety- related defect. Second, the RMA argues that disclosure should be consistent with the general treatment of information exchanged during discovery in private litigation as opposed to the mandates of the Freedom of Information Act. Both premises are wrong.

      \10\ Under Exemption 4, the information to be protected must be commercial or financial. Data relating to fatalities, injuries and property damage claims are based on certain information received involuntarily by the manufacturers, and do not constitute commercial or financial information. See generally, National Ass'n of Home Builders v. Norton, 309 F.3d 26, 38-39 (D.C. Cir. 2002).

      \11\ Under the early warning regulation, claim and notice information is different from customer complaint data. Customer complaints are communications received by manufacturers expressing dissatisfaction with a product (whether because of performance or the possible presence of a defect) and by definition do not include a claim or notice involving a fatality or injury. In general, claims involve written requests or demands for relief that a manufacturer receives, and notices refer to information received by a manufacturer (other than a media article), that do not include a demand for relief. Customer complaints reveal overall customer experience, while claims and notices reflect specific claims for relief premised on allegations of actual injury or damage.

      The final rule made clear that the purpose of the early warning data is to provide the agency with information indicating possible safety-related problems in motor vehicles and equipment. The data will assist the agency in determining what issues should be investigated to ensure that safety related defects are addressed expeditiously. Early warning

      [[Page 21415]]

      information, coupled with other information in the agency's possession, will be used to identify appropriate issues for investigation, and will not, in themselves, demonstrate that a safety-related defect exists.

      Like NHTSA, RMA is of the view that EWR data relating to fatalities, injuries and property damage are not defect data. RMA has not provided support for its premise that--contrary to its position-- these EWR data will be perceived as defect data. In any event, RMA proceeds to assert that the data should be treated confidentially because there will be cross-company comparisons. Even assuming cross- company comparisons based on the death, injury and property damage claims data could be made, the comparisons themselves do not give rise to substantial competitive harm within the meaning of Exemption 4. Nor has the RMA demonstrated that the comparisons would substantially affect purchasing decisions.

      The tire industry's market can be divided into two segments: sales of tires to vehicle manufacturers for new vehicles, which the RMA refers to as original equipment customers, and sales to the replacement market. Vehicle manufacturers are very sophisticated purchasers, and often are involved directly in the design of tires supplied by tire manufacturers. Vehicle manufacturers also have considerable experience with early warning data. The RMA has not shown that vehicle manufacturers would base tire purchases on the early warning death, injury and property damage claims information submitted by tire manufacturers, and the RMA's own statements that the information is not useful for comparisons (e.g., because of the absence of production data from which normalized rates could be developed) support our view that a competing tire manufacturer would not use the early warning claims data in a competitively harmful way.\12\

      \12\ This stands in contrast to the manner in which competitors could readily use other EWR data such as warranty data. For example, GM explained that warranty data provide an index of manufacturer costs and reveal a manufacturer's field experience with a particular component and supplier, which would enable competitors to benefit from a submitter manufacturer's experience in selecting suppliers. Disclosing warranty data would also benefit suppliers vis-a-vis a manufacturer since suppliers would receive information that they do not receive under current information sharing efforts. Similarly, JPMA observed that warranty data provide ``real time'' information concerning a company's production capacities, sales and market performance, which, if disclosed, would enable competitors to identify vulnerabilities and allow them to target production and marketing efforts accordingly. The effects of disclosure are discussed in greater detail in the section on warranties and in the final rule's preamble.

      Similarly, the RMA has not demonstrated that the release of these categories of early warning data likely would cause substantial competitive harm in the replacement tire market. As indicated by the RMA's petition, tire manufacturers themselves would not make, and would generally deny the validity of, any comparisons based on these data. This view of the validity of comparisons suggests that competitors would not go to the effort to develop comparisons and substantially undercuts their impact.

      Even assuming that someone would attempt to make a comparison based on death, injury and property damage claims information, the publicly available information is limited and not useful for comparisons, as recognized by the RMA. Tire manufacturers must provide separate reports by tire line, tire size, stock keeping unit, manufacturing plant and production period. 49 CFR 579.26. As a result, their reports will include numerous rows of data. In contrast, the numbers of incidents of deaths and injuries in the claims information submitted by tire manufacturers on December 1, 2003 are not numerous, particularly when compared to the numbers of sizes and models of tires.\13\ In addition, the absence of production data precludes the development of normalized rates (e.g., claims per 100,000 tires) that would be needed for comparisons.

      \13\ At the time that the RMA submitted its petition for reconsideration almost a full quarter of reportable early warning data was in the hands of the RMA's members.

      In any event, and perhaps because of these fundamental limitations, the RMA has not shown how the modest amount of data present in the submissions would be used in cross-company comparisons, who would perform them, or the competitive significance of those comparisons. Nor has the RMA addressed the fact that some smaller and lesser-known tire companies reported few to no death, injury, or property damage claims, which could readily prompt conclusions by potential consumers that it was not surprising that a small company received few claims and, therefore, that comparisons based on these early warning data do not substantially influence purchasing decisions.

      Nor do we find persuasive the RMA's suggestion that because information like the early warning data is often--but not always-- subject to protective orders in private litigation, it should be protected from disclosure under Exemption 4 of the Freedom of Information Act. Protective orders may be issued under a broad standard ``to protect a party or person from annoyance, embarrassment, oppression, or undue burden or expense.'' See Fed. R. Civ. P. 26(c). Exemption 4 is narrower, and the courts have recognized that the standards applied to protective orders and under FOIA differ. See Burka v. HHS, 87 F.3d 508, 517 (D.C. Cir. 1996); Anderson v. HHS, 907 F.2d 936, 946 (10th Cir. 1990).\14\

      \14\ Moreover, as a practical matter, protective orders often are submitted on consent by the parties in a civil action, and the court does not see the documents or independently assess the consequences of revealing them to non-parties. Thus, the fact that courts have issued protective orders is not particularly meaningful in determining the confidentiality of documents under Exemption 4.

      Finally, the RMA contends that its members' claims information form compilations of information that are confidential. As noted in the final rule, the collection of specific information that, when assembled together, would reveal sensitive commercial information can be protected under Exemption 4. For example, in Trans-Pacific Policing Agreement v. United States Customs Service, 177 F.3d 1022 (D.C. Cir. 1999), the court recognized that a compilation of complete commercial shipping code information that would reveal competitively sensitive information could be protected under Exemption 4.

      Not all compilations, however, meet these criteria. As noted above, the compilation of warranty, field report and consumer complaint data permits competitors to evaluate how consumers, suppliers and others in the market respond to various product-related decisions. Competitors may use the information (not otherwise available without substantial investment) to advance their own product offerings. The compilation of these categories of data reveals substantially more, both qualitatively and quantitatively, than the revelation of the various pieces of individual data.

      In contrast, the fatality, injury and property damage claims information is a collection of data points. While we believe these data will be useful in identifying possible safety problems worthy of investigation, the compilation of those data points does not confer competitive value to the data elements themselves. Nor does the RMA explain how competitors could use the claims information, whether individually or collectively, to advance their competitive advantage. Since the RMA's petition does not establish that disclosure of this early warning information will cause substantial

      [[Page 21416]]

      competitive harm, the agency will not alter its decision to release this information.\15\

      \15\ Nor has the RMA provided any new or convincing information suggesting that disclosure of claims information would impede the agency's defect program.

      2. VIN and State (or Foreign Country) Information

      The Alliance requests that the agency expand its early warning reporting class determinations in Appendix C to cover two items of information provided in reports of incidents involving fatalities or injuries. See generally, 49 CFR 579.21(b)(2). First, it asks that VIN information included in fatality and injury claims data be accorded confidentiality because the VIN could be used to trace the identity of the vehicle owner(s). The Alliance provided information showing that individuals can be easily identified by using VIN data as the starting point and coupling this information with information from commercially available databases.

      Second, the Alliance requests that the identification of the state or foreign country where the incident occurred be treated as confidential. The Alliance argues, without providing separate justification, that in sparsely populated states, an individual could research local media outlets to determine the identity of the individuals involved in the incident.

      We have decided to add to Appendix C the last six digits of the VIN data in the information on deaths and injuries. We have decided not to do so with regard to information relating to the state or foreign country in which incidents occurred.

      VIN Information. Each VIN consists of 17 characters. In general, the first eight characterize the manufacturer and attributes of the vehicle including the make and type of vehicle (e.g., the relevant line, series, body, type, model year, engine type and weight rating). The ninth digit is a check digit. In the last eight characters, the first two represent the vehicle model year and plant of production, and the last six are the number sequentially assigned by the manufacturer in the production process. See 49 CFR 565.6 (detailing elements of the VIN code), SAE Standards J218 (passenger car identification terminology) and J272 (vehicle identification number systems).

      Under the final CBI rule, NHTSA's disclosure of fatality and injury data included the entire VIN. Based in part on our consideration of the Alliance's petition for reconsideration, we have decided to modify the rule to disclose the initial 11 characters of VINs and hold the remaining 6 characters confidential. The disclosure of the initial 11 characters provides information on the vehicle identified in the claim or notice, beyond make and model information that is already available. See, e.g. 49 CFR 579.21(b)(2); see also 68 FR 44221-22. The release of this VIN information is not accompanied by a risk of violating an owner's privacy.\16\ However, based in part on the Alliance's petition for reconsideration, we will hold confidential the last six characters of the VIN because they can be used to obtain personal identifying information.

      \16\ This redaction policy is consistent with the one followed by our Office of Special Crash Investigations.

      Following review of the Alliance's petition, the agency examined a widely available legal database--WESTLAW--and several websites that offered to provide personal information on individuals using the VIN of a vehicle for a nominal fee. The agency was readily able to determine the name, address, date of birth, and lien information of the vehicle owner using the full VIN. In view of this easy identification, the disclosure of full VIN information would jeopardize the personal privacy of individuals involved in EWR reports of fatalities and injuries arising from motor vehicle crashes.

      NHTSA is according confidentiality to the last six digits of VINs under FOIA Exemption 6, which protects personal privacy interests. Under Exemption 6, an agency engages in a balancing process. The first step in the process is an assessment of the privacy interests at stake. In Center for Auto Safety v. NHTSA, 809 F. Supp. 148 (D.D.C. 1993), the court recognized the privacy interests in the names and addresses on consumer complaints received by NHTSA. The same interests exist here. The second step is an assessment of the public interest. Under Exemption 6 the concept of public interest is limited to shedding light on the government's performance of its statutory duties. We note that the public will be able to review EWR information on claims for fatalities and injuries, including identification of the make, model and model year of the vehicle and the component or system implicated in the claim. Disclosing additional VIN information that would enable someone to identify the owner of the vehicle does not serve a public purpose. If disclosed, it would not answer the question of ``what the government is up to.'' Dep't of Justice v. Reporters Comm. for Freedom of Press, 489 U.S. 749, 773 (1989). See also National Ass'n of Retired Fed. Employees v. Horner, 879 F.2d 873, 879 (DC Cir. 1989) (the sought information must enable ``the public [to] learn something directly about the workings of the Government'') (emphasis in original). The final step in applying Exemption 6 is weighing the competing privacy and public interests against one another. As in Center for Auto Safety, the privacy of the persons who may be identified from the last six digits of VINs will be recognized and protected because there is no ascertainable public interest of sufficient significance or certainty to outweigh that right. 809 F. Supp. at 150.

      State and Foreign Country Information. We are denying that portion of the Alliance petition requesting protection for information relating to the state or foreign country in which incidents occurred. According to the Alliance, information relating to the location of an incident may allow interested parties to discover personal information about victims by perusing local newspapers or other reports relating to the event. While it is possible for the EWR information to be linked to other publicly available information, we do not believe that privacy interests are sufficiently jeopardized to justify withholding such information. As pointed out by the Alliance petition, the incidents of concern have already received some public attention and, therefore, personal information about those involved is likely already known on a local or state level. The disclosure of this information in the EWR reports is unlikely to shed much additional information into the public domain. 3. Common Green Tires

      The RMA asks the agency to clarify its position with regard to information submitted relating to ``common green'' tires and to create a class determination covering that reporting requirement. It notes that this particular issue was not addressed in the final CBI rule.

      The term ``common green'' refers to a basic tire construction used as the foundation for an array of different tire models and/or brands. This basic tire envelope is placed into different tire molds in the tire production process and serves as the foundation for tires with different tread patterns and different brand names. The early warning final rule defines ``common greens'' as tires ``that are produced to the same internal specifications but that have, or may have, different external characteristics and may be sold under different tire line names.'' 49 CFR 579.4(c). The early warning regulations

      [[Page 21417]]

      include a separate common green tire reporting requirement. As part of each quarterly report, each manufacturer of tires provides NHTSA with a list of common green tires and for each specific common green tire grouping, the listing includes relevant tire lines, tire type codes, stock keeping unit (``SKU'') numbers, brand names, and brand name owners. 49 CFR 579.26(d).

      The RMA explains that the disclosure of common green tire listings is likely to cause competitive harm because common greens reveal the relationships between tire groupings, providing competitors with the ability to determine a tire manufacturer's marketing and business plans and potentially its cost structures. Because common green information would reveal the identities of tires that have the same internal specifications, as well as the relationships between manufacturers and private brand name owners, the RMA argues that the disclosure of this information would cause tire manufacturers substantial competitive harm.\17\

      \17\ As explained in tire manufacturer letters seeking confidential status for common green lists submitted on December 1, 2003, the disclosure of EWR ``common green'' information would allow competitors to assess a tire manufacturer's technical capabilities and marketing strategies, as well as tire production and mold design technology. In addition, competitors could use the ``common green'' information to determine a tire manufacturer's future product plans.

      A report submitted by Professor Michael D. Bradley that accompanied Cooper Tire's comments to the docket notes that common green tire information serves as the basis for tire line production and that the release of this type of information would provide a ``complete and comprehensive'' picture of a tire company's production and marketing strategies. The report observes that the disclosure of this information would be equivalent to the release of a tire company's business plan.

      The agency did not specifically address ``common greens'' in the final CBI rule. We agree with the RMA that disclosure of ``common green'' information is likely to cause substantial competitive harm. The disclosure of ``common green'' lists would reveal to competitors a tire manufacturer's production strategies, marketing strategies, future product plans, and its tire production and mold design approach.

      In addition, ``common green'' identifier data are appropriate for a class determination under Appendix C. The common green is the basic envelope of tire production. The type of ``common green'' identifier information submitted under the EWR rule is the same for all manufacturers. The impact such information would have in the competitive market will not vary and individual consideration of each submission would result in identical determinations that disclosure is likely to lead to substantial competitive harm. Accordingly, we have added a class determination to Appendix C covering the submission of ``common green'' identifier data pursuant to the early warning regulation. 4. Production, Warranty Claims and Adjustments, Field Reports and Consumer Complaints

      PCLG takes issue with the agency's class determination of confidentiality of EWR production data (for all products other than light vehicles), warranty claims information, field reports and consumer complaints. After reviewing the comments and the applicable law, the agency determined that release of this information was likely to cause substantial competitive harm and to impede the agency's early warning program. PCLG argues that all of this information should be publicly available.

      In reaching its determination, the agency balanced private and public interests consistent with Exemption 4. NHTSA considered the manufacturers' interest in legitimate protection from competitive harm as specified by Exemption 4 of FOIA and balanced the various public policy issues involved--the public's interest in disclosure and the extent of impairment that likely would follow from disclosure.

      PCLG asserts that the agency improperly applied these policy considerations, arguing principally that the information collected under the early warning program should be disclosed to allow the public, as well as the agency, to assess whether potential safety related defects exist. PCLG also asserts that the agency should not hold early warning information confidential without individual consideration of each manufacturer's competitive situation and whether disclosure will likely cause substantial competitive harm to that manufacturer. Similarly, PCLG asserts that in such reviews the agency should segregate any portion of the early warning data that will not cause competitive harm or impair the government's program. Finally, PCLG takes issue with the agency's application of the impairment prong of Exemption 4.

      PCLG's objection to the agency's approach closely parallels its argument that the agency lacks the legal authority to establish class determinations. PCLG advances the proposition that the agency must make individual decisions with regard to individual submissions of EWR data under FOIA.\18\ It also observes that the agency makes some information submitted by manufacturers in individual investigations of alleged defects public and disagrees with the agency's determination that the comprehensive compilation of early warning information is quantitatively and qualitatively different from the specific data provided by manufacturers in response to NHTSA Office of Defects Investigation (``ODI'') information requests in particular defect investigations.

      \18\ While PCLG offered several cases to support its view, the cases did not support this broad proposition. The cases fairly point out that there needs to be an adequate basis for withholding information. NHTSA believes that there is adequate support for Appendix C to the CBI rule.

      As noted above, the early warning program is a unique government program. The information is being collected and analyzed electronically. Unlike most government programs, much of the data will never be directly relevant to any particular agency investigation or regulatory activity. The agency is unaware of any similar government database. In this context, cases reviewing particular competitive assessments under particular facts in light of a particular submission to the government shed no light on whether the wholesale disclosure of business information is likely to cause substantial competitive harm.

      Nor is this a situation, as with individual submissions like those in the course of defect investigations, that allows for the segregation of data beyond the categorizations the agency has already applied. The agency has already segregated the early warning data into the various categories of information to be provided and, as set forth in the final rule, considered each category separately. As a result, the agency determined that while some categories of early warning data were entitled to confidential treatment, other categories should not be. We do not believe it is possible to further segregate the data within each category, as each category contains from each manufacturer the same type of data presented in a required format. The early warning database is fundamentally different than individual submissions (such has those presented during defect investigations) in which confidential data is routinely redacted and the remainder of the submission is placed in the public file.

      PCLG takes issue with the agency's consideration of the potential of the release of comprehensive early warning data to cause competitive harm. PCLG challenges the notion that cross-

      [[Page 21418]]

      company comparisons that may arise from disclosure are of a nature falling within the purview of Exemption 4. According to PCLG, such analyses merely reveal the performance of products and present the public with the same kind of information available through popular magazines and from other public sources.

      PCLG submitted an Appendix to its petition for reconsideration containing numerous newspaper articles and other publicly available sources making cross-company comparisons or discussing vehicle performance issues. PCLG asserts that such information negates the agency's finding that disclosure of the compilation of data relating to raw warranty claim and consumer complaint information and field reports is likely to cause substantial competitive harm.

      The Appendix is not persuasive. The types of reports included in the Appendix are the result of largely anecdotal reports or limited data collection efforts. The public sources of information do not appear to remotely resemble, in breadth, subcategorization or objective underpinnings, the EWR warranty data. Unlike EWR data, they do not carry the imprimatur of being a comprehensive set of data collected directly by the manufacturers and submitted to the government for analytic review. Nor does PCLG's position consider the agency's concern, based on the comments in the docket, that disclosure will lead manufacturers to provide less warranty coverage, conduct fewer internal investigations and put fewer resources towards the receipt and resolution of consumer complaints. The result will be less information available to the agency's early warning detection program.

      While we do not find the general arguments persuasive, below we review each category of confidential information in response to the assertion that the agency lacked information from which it could make a determination whether disclosure likely would cause substantial competitive harm to the submitter of the EWR information or impair the government's program.

      1. Production Numbers. The final rule created a class determination for reports of production numbers for medium-heavy vehicles including buses, motorcycles, trailers, child restraint systems and tires--i.e., the manufacturers covered by the comprehensive EWR requirements, except for light vehicle manufacturers. Light vehicle production is reported publicly. As we explained:

         Many business interests discussed their efforts to maintain the confidentiality of their production figures. Harley-Davidson and MIC

         [the Motorcycle Industry Council] stated that production numbers by model have never been generally available in the motorcycle industry. Cooper Tire submitted an affidavit, further confirmed through RMA's comments, with regard to the competitive harm that disclosure of otherwise confidential production numbers would have in the tire industry. JPMA argued that disclosure of these data would provide new entrants and competitors in the child restraint industry with information about production capacities, sales and market performance not otherwise available in the absence of considerable investment in market research. Bluebird (busses, school buses and motor homes), Utilimaster (final stage walk-in vans and freight bodies for commercial use) and the AORC (occupant restraint systems and other components) also each stated that production numbers in their segment of the industry are confidential and likely to lead to substantial competitive harm if released.

         The comments substantiate that production numbers in many sectors of the automotive and equipment industries are competitively protected information, revealing otherwise unobtainable data relating to business practices and marketing strategies. [68 FR 44221]

         The record amply supports the creation of a class determination on the production numbers for vehicles and equipment subject to EWR reporting, other than light vehicles. Production numbers from these other sectors are competitively sensitive data. For example, RMA explained that tire production numbers, which are reported by (among others) tire line, tire size, stock keeping unit and plant of production (49 CFR 579.26), were competitively sensitive and that their routine disclosure to the public through EWR submissions would, among other things, enable competitors to analyze their competitors' businesses. Cooper Tire, noting the competitively harmful effects that would accompany the disclosure of production data, emphasized that the intense level of competition within the tire industry and the size differences among competitors made the risk of substantial competitive harm high, particularly for smaller tire manufacturers that produce products for the replacement market. An accompanying economist's report noted that the tire industry is ``highly concentrated'' and that the disclosure of production numbers would reveal substantial information related to company marketing plans and strategies.

         PCLG's specific objections to the class determination for production levels rest on its broad assertions that the class determination is not supported by the factual record and is inconsistent with the agency's past practice to disclose production information. However, PCLG's petition neither addresses the record nor provides factual or expert rebuttal.

         PCLG further asserts that the agency's past practice has been to treat production numbers as confidential and provided an example to support its assertion. PCLG's assertion is inaccurate and is not supported by its example. As stated in the final rule, production numbers for manufacturers other than light vehicle manufacturers have been treated confidentially in the past on the basis that their disclosure is likely to cause substantial competitive harm to businesses engaged in these industries. 68 FR at 44221. Such an example is found in the Closing Report in ODI's investigation of certain Goodyear tires (PE 00-046). PCLG's example of past disclosure practices amounts to the release of warranty data, rather than production data, during the course of one defect investigation. This is of no bearing because investigation information is not comparable to that submitted under the EWR rule and because the confidentiality of warranty data is not determinative of the confidentiality of production data.

         PCLG's other and more generalized arguments do not require a different result. PCLG's arguments disputing the bases for the class determinations under Exemption 4--i.e., unwarranted product disparagement and competitor usage of data--do not refer or apply to production numbers.

      2. Warranty Claims and Adjustments. The final rule established a class determination for warranty claim numbers for vehicles and child restraint systems, and for warranty adjustments in the tire industry. As noted in the preamble, the disclosure of warranty data is likely to cause substantial competitive harm:

         [T]he warranty information required by the early warning reporting rule--that is, the number of claims associated with specific components and systems broken down by make, model and model year (with slightly different breakouts for tires and child restraint systems)--is likely to provide competing manufacturers with sufficient information about the field experience of those components and systems to provide commercial value to competitors who may be deciding whether to purchase similar components, the price at which to purchase those components and which suppliers to choose. * * *

         While manufacturers are likely to explore the practices and policies of their competitors when reviewing any publicly available warranty claims information, the public is more likely simply to rely on generic cross-company comparisons. The warranty claims information may be used as

         [[Page 21419]]

         part of vehicle comparisons, even though the warranty terms and conditions and corporate warranty practices may differ. As a result, the potential for the warranty claims information to give rise to misleading comparisons and cause substantial competitive harm is also strong. * * * * *

         [W]e have determined that the early warning reporting of warranty information . . . is entitled to confidential treatment because its disclosure is likely to cause substantial competitive harm.

         The warranty data required by the early warning reporting regulation are also entitled to confidential treatment because their disclosure is likely to impair the agency's ability in the future to obtain and use reliable warranty information as part of its program to identify potential safety related defects. Warranty claims data-- which begin to accumulate as soon as vehicles are sold and continue for the length of any given warranty policy--will be a significant indicant of potential defects. The quarterly warranty claims reports, combined initially with the historical seeding material, will help the agency to identify trends involving particular equipment and systems or components in a particular make, model and model year of a given product. [68 FR 44222-23]

         The record supports NHTSA's conclusion that warranty data have a variety of direct competitive uses. For example, the Alliance, whose members produce light motor vehicles, medium-heavy motor vehicles and motorcycles, through a report prepared by AutoPacific, explained that:

         Actual working experience at various automotive companies confirms that comparative component warranty experience, reliability experience, and durability experience strongly influences component pricing and sourcing decisions. . . . [I]f one original equipment manufacturer purchases a component and obtains field experience with that component, it can be expected to use that information to make decisions about future purchases and the price it will pay. Providing that field experience to other manufacturers effectively gives them a ``free ride'' at the expense of the first manufacturer.

         [Comments, Attachment A, at 1]

         The Alliance's comments further noted the particular value that EWR warranty data have by detailing the manner in which they may be employed both by current and potential competitors who may decide to enter into the U.S. market. In addition, GM explained that warranty data provide an index of manufacturer costs. An article referenced in the preamble to the final rule described the direct use of warranty data by manufacturers to help them analyze and identify problems encountered in their vehicle fleets. Gregory White, ``GM Takes Advice from Disease Sleuths to Debug Cars,'' Wall Street J., April 8, 1999 at B1 (describing GM plan to use warranty data to detect vehicle problems and eliminate claims and noting statistical analysis employed by rival DaimlerChrysler to accomplish the same).

         The TMA, representing medium and heavy truck manufacturers, explained that the disclosure of comprehensive warranty data such as that collected under the EWR rule would provide competitors with previously unavailable market intelligence that is of much greater breadth and depth than the information contained in typical information submissions to the agency. RMA expressed concerns that the disclosure of warranty adjustment data would reveal different policies among tire manufacturers. Similarly, JPMA explained that competing child restraint manufacturers could use this information to their advantage and to the detriment of the submitter and it stressed that the data would provide competitors with real time, ongoing competitive information on a company's production capacities, sales and marketing performance.

         PCLG's petition does not rebut the factual premises in the record of NHTSA's class determination on warranty claims, which includes warranty adjustments in the tire industry. Instead, PCLG attempts to make a case in favor of disclosure by submitting information on the agency's determination that certain information supplied by a vehicle manufacturer within the context of a specific investigation by ODI was not confidential. That sample submission, however, does not involve or represent EWR information.

         As discussed at length in the final CBI rule, there are substantial differences between data submitted pursuant to the EWR rule, which contain information about the entire product lines of a manufacturer, and the limited and narrow information submitted by a manufacturer in response to an agency information request issued during the course of an ODI investigation. PCLG does not address or deny these inherent differences.

         More generally, PCLG advances a number of arguments to support its view that EWR warranty numbers are not confidential. As one broad theme, PCLG disputes that a competitor's use of EWR warranty data to assess field experience for purposes such as durability assessments, purchasing, pricing and supply decisions, which PCLG calls avoiding development costs, is an adequate basis for treating these warranty data as confidential. In support of its position, PCLG asserts in part that competitive harm cannot be based on possible competitor use of data that identify safety problems in vehicles on the market. Both the implicit factual premise for this assertion and the legal basis for it are unfounded. Factually, EWR data are not data on safety problems. EWR warranty data reflect payment of warranty claims involving various systems and components, such as the power train and seat, without the identification of any particular component or any problem. The fact that a manufacturer made warranty payments for vehicles, tires or child restraints does not mean that these products contain safety-related defects. NHTSA's consideration of the data in the early warning review process does not suggest otherwise. NHTSA is using raw EWR warranty claims data as a tool in assessing whether a defect potentially exists. See 67 FR 45852 (July 10, 2002). The agency is reviewing these data for trends. Most data are not likely to indicate a potentially problematic trend. As to data appearing to indicate possible trends, ODI may make inquiries to manufacturers. If the agency's assessment of all available information, including (where appropriate) the manufacturer's response to its inquiries, indicates that an investigation is warranted, the agency will open an investigation. 67 FR at 45865. Thus, contrary to PCLG's suggestion, EWR warranty data do not in themselves identify safety problems.

         Legally, PCLG contends that the EWR warranty information is being withheld to protect the manufacturer's reputation or ability to continue to sell the equipment. PCLG argues that the harm resulting from such disclosures is not a cognizable competitive harm under Exemption 4 and that revealing safety problems does not result in an unfair advantage to competitors, citing Public Citizen Health Research Group v. FDA, 704 F.2d 1280 (D.C. Cir. 1983). In that case, the court remarked in a footnote that competitive harm in the FOIA context is limited to the harm flowing from the affirmative use of proprietary information by competitors and that competitive harm should not be taken to mean simply any injury to competitive position as might flow from embarrassing publicity attendant upon public revelations concerning, among others, violations of safety laws.

         PCLG's reliance on that proposition is misplaced. The EWR warranty data are being withheld because of the competitive harm that likely would flow from their disclosure, as discussed above, and not because of concerns over the manufacturer's reputation or ability to continue to sell the equipment. See 68 FR 44222-23. In view of the

         [[Page 21420]]

         competitively sensitive nature of the data, under Exemption 4 the data are confidential notwithstanding that withholding them would, as viewed by PCLG, save the manufacturer from the noncognizable harm of embarrassment. This conclusion is fully supported by a case that specifically clarified Public Citizen. In Occidental Petroleum v. SEC, 873 F.2d 325, 341 (D.C. Cir. 1989), the court made clear that the possibility that some noncognizable harm would flow from disclosure is not dispositive under Exemption 4, since an agency's role is to determine whether non-public information contained in documents is competitively sensitive, for whatever reasons.

         PCLG also asserts that the release of the EWR warranty claim counts would provide significant information about component performance only in extreme cases and concluded that competitive injury supporting a class determination would rarely arise. We disagree. As pointed out in the Alliance's comments, these data are competitively sensitive on concerns such as component and system performance and reliability. Competitors and potential consumers would utilize these data, regardless of whether they reflect potential problems, the likelihood of few problems or otherwise. The fact that consumers would use the information to make comparisons, as a supplement to other sources of comparative information in purchasing decisions, was recognized in PCLG's comments. It simply ignored the fact that manufacturers could use the same information to the disadvantage of a competitor in the manner described above.

         In the alternative, PCLG observes that even if the data reveal competitively valuable and sensitive information on good or bad performance, they may be a matter of public knowledge, there may be no competition on a system or component and there must be a showing that the competitors could not obtain the information at a reasonable cost. Whatever public knowledge there may be about a problem (e.g., press anecdotes), as discussed above it would not be comparable to the EWR warranty claim submissions. Moreover, the motor vehicle, equipment, tire and child restraint businesses are highly competitive, and the record shows that the EWR warranty data are not available and would not be available at a reasonable cost. PCLG does not show otherwise.

         PCLG adds that the information must have commercial value and argues that information related to components uniquely suited to a particular vehicle could not have competitive value. However, the value of the information is not dependent on whether a specific component has a single or multiple vehicle applications. EWR information provides insight into a broad range of issues, including field experience, customer satisfaction and cost decisions made by companies in paying warranty claims. This is a type of information that Exemption 4 was designed to enable the government to protect.

         In another broad theme, PCLG asserts that NHTSA based its class determination on the noncognizable harm of unwarranted product disparagement arising from misleading company comparisons of warranty claims information. In the preamble to the final CBI rule, we recognized that warranty claims information may be used as part of vehicle comparisons, even though warranty terms and corporate warranty policies may differ, resulting in a strong potential for warranty claims information to give rise to misleading comparisons and cause substantial competitive harm. See 68 FR 44222-23. PCLG requests reconsideration of NHTSA's conclusions that the use of cross-company comparisons could result in substantial competitive harm.

         First, PCLG asserts that NHTSA ignored well-established data sources, such as Consumer Reports, which is available to consumers seeking to make cross-company comparisons. It contends that consumers would treat the EWR warranty data as another source of information and that professionals would recognize the limitations of the data and evaluate them in context. In light of these other sources of information, PCLG discounts the competitive effect of release of the information. However, PCLG does not identify the comparisons that would be made using existing publicly available information or establish the comparability of public data to EWR warranty data. As discussed above, it appears that the public sources of information do not remotely resemble the EWR warranty data. Accordingly, we do not accept PCLG's theory.

         Second, PCLG asserts that insofar as the rule was based on competitors' use of their rivals' information to make misleading comparisons and engage in unwarranted product disparagement, these comparisons are not a proper ground for withholding EWR data under Exemption 4. PCLG notes that laws preclude misleading marketing and the impact from misleading marketing will not be so widespread as to result in significant competitive harm. We believe that PCLG misunderstood NHTSA's rationale. The agency based the rule in part on the competitive harm that flows from the use of EWR warranty data by competitors and by consumers. We did not base it on misleading and unlawful product disparagement by competitors.

         Third, PCLG argues that the possibility that information may be misinterpreted has never been recognized as a justification for according confidentiality to information. It notes that virtually all data can be misinterpreted and data cannot be withheld on this basis. However, in Worthington Compressors, Inc. v. Costle, 662 F.2d 45, 53 (D.C. Cir. 1981), the court permitted the consideration of consumer misuse of commercial information that is otherwise unavailable. Accordingly, NHTSA was authorized to treat EWR warranty data as confidential on this alternative basis.

         In its final assertion on unwarranted product disparagement, PCLG contends that the harm occurring from the disclosure of these data amounts to adverse public reaction, which is not a cognizable harm under Exemption 4. See Public Citizen Health Research Group v. FDA, 964 F. Supp. 413, 415 n.2 (D.D.C. 1997). Factually, we adhere to our views of the harm as stated in the preamble to the final rule and disagree with PCLG's attempt to recharacterize the harm and eliminate the harms we identified. Since the EWR warranty data are competitively sensitive for a valid reason under Exemption 4, other potential consequences such as adverse public reaction, do not dictate that we treat the information as non-confidential. Occidental Petroleum v. SEC. The final CBI rule is based on such valid determinations, as described in the preamble.

         PCLG's third broad theme is that the agency did not satisfy the impairment prong of Exemption 4 in its assessment of the release of EWR warranty data. Under the impairment prong, an agency may withhold information that, if released, ``would impair the effectiveness of a government program.'' Public Citizen v. NIH, 209 F. Supp. 2d 37, 52 (D.D.C. 2002). See also 9 to 5 Organization for Women Office Workers v. Federal Reserve System, 721 F.2d 1, 11 (1st Cir. 1983), and Appendix B to the final rule's preamble.

         NHTSA carefully considered the value of warranty claim data to the defect identification program and the impact that disclosure would have on manufacturer policies and decided that EWR warranty data should not be disclosed. The importance of warranty information had been explained:

         [[Page 21421]]

         We have often found warranty claims to be more valuable than customer complaints because the customer has identified a problem, a repair facility . . . has performed a repair, and the manufacturer has paid for some of or all the repair. This information is valuable to NHTSA as an early warning tool in assessing whether a defect potentially exists. The principal limit on the value is that after the expiration of the warranty . . ., this information is no longer generated. However, at times these programs are extended when there are problems with the product and at times manufacturers also pay for repairs under ``good will'' programs. We have found that ``good will'' actions provide valuable information in that manufacturers may choose to address a perceived problem by extending or liberalizing the terms of a warranty rather than by conducting a full recall, or by formally extending the warranty period. In order to aid in the early discovery of potential defects, the agency believes that the number of good will claims should be reported along with the more ``traditional'' warranty claims. [67 FR 45852 (July 10, 2002)]

         Manufacturers with generous warranty or good will programs will have a higher number of warranty claims than they would have with less generous programs, and releasing these data would create the perception that these manufacturers' products have relatively more problems.\19\ Disclosure would encourage manufacturers to restrict more generous warranty and good will programs in order to report lower warranty numbers data. The restriction of warranty programs and consequent reduced reporting will reduce the amount of warranty information that the agency may consider. This would impair the agency's ability to determine whether a defect trend in a particular line of vehicles, equipment or tires exists, as well as potentially increasing the inconvenience to consumers. 68 FR at 44222-23. These effects are supported by comments in the record, including those from the Alliance, the Tire Industry Association (``TIA''), the Association of International Automobile Manufacturers (``AIAM''), and Workhorse.

         \19\ The term ``good will'' refers to those repairs that are ``paid for by the manufacturer, at least in part, when the repair or replacement is not covered under warranty, or under safety recall reported to NHTSA under part 573 of [Chapter 49].'' 49 CFR 579.4(c).

         PCLG asserts that the TREAD Act requires the submission of EWR data, which makes NHTSA's claim of impairment difficult to justify. This comment misses a critical underpinning of EWR reporting. While the TREAD Act authorizes the agency to compel manufacturers to provide data that they already collect, it explicitly precludes NHTSA from requiring the submission of information not in the possession of the manufacturer. 49 U.S.C. 30166(m)(4)(B).\20\ It also does not authorize NHTSA to require good will repairs and does not restrict a manufacturer's discretion to set or reduce warranty coverage or good will repairs.

         \20\ This section provides that the EWR regulations ``may not require a manufacturer of a motor vehicle or motor vehicle equipment to maintain or submit records respecting information not in the possession of the manufacturer.''

         PCLG contends that other factors may influence a manufacturer's decision to provide extensive warranties, making the likelihood of impairment remote. We recognize that customer-oriented factors have a significant influence on the scope and extent of warranty programs. However, we agree with the manufacturers that publication of the EWR data would give some manufacturers ``black eyes'' and that to a notable degree it is likely they would alleviate this problem, and improve sales and profits, by limiting warranty coverage, including good will payments. This would reduce the numbers of claims in the EWR warranty database, particularly toward the end of a warranty period and beyond, when components often break. ODI's analysis of warranty data to identify possible defects, which is predicated on substantial numbers to detect trends, would be impaired, as would its use in defect investigations. The agency thus believes that the risk of impairment associated with the wholesale disclosure of information such as warranty data is sufficient to justify the agency's application of the impairment prong of Exemption 4. Information in the record adequately supports NHTSA's conclusion. See, e.g. Comments from the Alliance, TIA, AIAM, and Workhorse.

         PCLG adds that under the impairment prong there must be a rough balancing of the importance of the information and the extent of the impairment against the public interest in disclosure. See Washington Post v. HHS, 690 F.2d 252, 269 (D.C. Cir. 1982); Washington Post v. HHS, 865 F.2d 320, 326-27 (D.C. Cir. 1989). It states that this balancing is not in the record.

         The importance of warranty claims data is explained in the record. The customer has identified a problem, a repair facility has performed a repair, and the manufacturer has paid for some of or all the repair during or after the warranty period. Separately, by model and model year, the numbers of warranty claims, by system and component, are reported to NHTSA. The magnitude of the numbers is important to the agency, as in our screening we will look for trends based in part on relatively high numbers. We believe that if warranty data were disclosed, given manufacturers' ability to set warranty coverage and to authorize good will repairs, warranty and good will coverage would be reduced and our ability to detect potential problems would be diminished. The resulting impairment to NHTSA would be substantial.

         On the other hand, the public interest in disclosure of warranty information is limited. Warranty data are simply payment data. Standing alone, the EWR data simply provide numbers of warranty claims payments, by system or component. They do not identify the particular part or a problem. Based on EWR warranty data alone, we believe it is not possible to accurately identify a safety-related defect in a particular product.

         PCLG also argues that data pertaining to older products, and by extension older technology, cannot qualify for protection under Exemption 4. This argument, however, ignores the baseline, competitive value of older data. For example, older information forms useful baselines for comparisons, which can be valuable in evaluating whether new technology is more durable than older technology. PCLG's argument also ignores the impairment concerns we identified in the final rule. Consequently, the agency believes that applying the class determinations set forth in Appendix C to older data continues to have merit.

         Finally, the competitive value of these data as a whole, for numerous separate reasons discussed above, also resolves the issues raised by PCLG on segregating EWR warranty data. The data cannot be segregated without revealing competitively sensitive information. PCLG offers no suggestions on how these or other EWR data could be segregated to avoid the concerns we identified in the preamble to the final rule and above.

      3. Field Reports. The final rule created a class determination of confidentiality for EWR field reports. Under the EWR rule, certain vehicle and child restraint systems manufacturers must report the total number of field reports they receive from their employees and representatives, and from dealers and fleets, that are related to problems with certain specified components and systems. In addition, these manufacturers must submit copies of field reports, except those received from dealers.

         [[Page 21422]]

         As explained in the preamble to the final CBI rule, field reports reflect the in-use experience of a manufacturer's product, identifying specific problems encountered in the field not otherwise available to competitors. This information allows manufacturers to conform future design and production to field experience. Because the disclosure of this information would enable a company to improve its products without the need to invest in market research, engineering development or actual market experience, these data have substantial commercial value. Release of this information would reveal to competitors product features, components and systems which have met with consumer acceptance (and which have not) as well as what problems may be associated with certain components and systems. Using field reporting data and reports themselves, a competitor may determine areas of importance to a manufacturer (whether potentially related to safety or not) and enable a competitor to note the expected experience from a particular component or system.

         The record supports the confidentiality of field report information. For example, the Alliance stated that a wholesale disclosure of field report data would enable industry-wide comparisons of component performances. AIAM noted that field reports would provide useful information on the manufacturing processes and cost structures to competitors without having to conduct the research to develop the information independently. The Truck Manufacturers Association (``TMA'') stated that disclosing field report data would reveal unobtainable market intelligence about a manufacturer and the operational status of its customers' vehicle fleets. Blue Bird indicated that field report data would assist competitors in conducting market research and strategic planning. It emphasized that the disclosure of these reports could compromise customer fleet operations. Although it generally questions the sufficiency of the record, PCLG does not address this or other record information in its petition.

         In addition to causing substantial competitive harm, it is likely that the disclosure of field report information would reduce the field report data received by NHTSA, both in terms of the number of reports and their depth of content. 68 FR 44224. Comments in the record bear this out. TMA stated that disclosure of field report information would likely lead to the creation of fewer and less informative field reports and a consequent reduction in the quality of information submitted to NHTSA. Similarly, the AIAM expressed concern about diminished thoroughness and candor if they are disclosed to the public. Blue Bird stated that NHTSA can reasonably anticipate that manufacturers would take measures to minimize field report information if disclosed. This record information supports NHTSA's conclusion that under the impairment prong of Exemption 4, the agency may hold field report information confidential to ensure the quantity and quality of information it receives during the EWR process.

         In general, PCLG's petition mentions field reports along with warranty claims, without a particular discussion of field reports. See PCLG Petition for Reconsideration at 7, 9-10. Accordingly, in response, we refer the reader to the discussion pertaining to EWR warranty claims above. In addition, the following supplements the discussion above, with regard to field reports.

         NHTSA's ODI has reviewed numerous field reports over the years. While they vary considerably in nature and quality, we often have found manufacturer field reports to be technically rich, although some, particularly by dealers, are less so. See 67 FR 45856. NHTSA also has held numerous field reports obtained in investigations confidential.

         Like EWR warranty claim data, field report data are not safety data. Field reports include reports on possible problems. However, the problems may merely be alleged by an owner of a vehicle or may be real. The perceived or actual problems addressed may involve performance that does not meet the expectations of the owner, but may not be significant. They may or may not be safety-related.

         NHTSA also balanced the importance of field reports and the extent of the impairment to the government against the public interest in disclosure. The importance of field reports is well established. By definition, an alleged failure, malfunction, lack of durability or other performance problem has been identified in a written communication to the manufacturer from one of its employees, representatives, dealers, or a fleet. 49 CFR 579.4(c). Under the EWR reporting program, the numbers of field reports, separately, by model and model year, and by system and component, are reported to NHTSA. The magnitude of the numbers of field reports is important to us, as in our screening we will look for trends based in part on relatively high numbers. These trend may result in inquiries to the manufacturers. We believe that, given manufacturers' substantial control over the direction of field activity and the preparation of field reports, if the numbers of field reports were disclosed to the public, the numbers of field reports would be reduced considerably and, as a consequence, our ability to detect potential problems would be highly diminished, causing a substantial impairment to the agency.

         On the other hand, the public interest in disclosure of field report numbers is limited. Standing alone, the EWR field report numbers simply indicate that there was a reported problem, by system or component. They do not identify the particular part or a problem. Based on EWR data alone, it is not possible to accurately identify a safety problem. Given these limitations, the public interest in disclosure is small. Thus, the impairment prong balancing weighs in favor of nondisclosure of field report data.

         The field reports themselves are very important to the government. They provide text that is not conveyed by the numerical reports. The views of manufacturers' engineers in reports are often helpful to us. If they were disclosed, manufacturers would react by decreasing the number of reports generated and the level of detail contained in these reports. Without them, we often would not gain a full understanding of the issue, at least not without a steep and time-consuming learning curve. We recognize that some of the field reports would be of interest to some members of the public. On balance, we are in a better position to address potential defects with as robust a set of field reports as possible, which benefits the public at large. Accordingly, NHTSA is justified in withholding EWR field reports under the impairment prong.

      4. Consumer Complaints. The final CBI rule created a class determination of confidentiality covering EWR consumer complaints. These include communications from consumers that express dissatisfaction with a product, note any actual or potential defect or any event allegedly caused by an actual or potential defect, or that relate to that product's unsatisfactory performance but exclude claims or notices involving a fatality or injury. 49 CFR 579.4(c).

         Consumer complaints provide information on the performance of products based on consumer feedback. They reveal which product features, components and systems have met with consumer acceptance (and which have not) and what perceived problems may be associated with particular components and systems. As noted in the preamble to the final CBI rule, the collection of consumer complaint data

         [[Page 21423]]

         is subject to company policies. For example, Harley-Davidson stated that it aggressively seeks consumer feedback while others may seek it but to a lesser degree. AIAM stated similarly that manufacturers may have consumer complaint processes that vary in efficiency.\21\

         \21\ The commercial value of consumer complaint data is well recognized. See e.g., Edward Bond & Ross Fink, Meeting the Customer Satisfaction Challenge, 43 Industrial Management, Issue 4 (July 1, 2001) (noting the importance of measuring customer satisfaction, describing customer complaints as a data source to a company that can create a ``big benefit'' from small changes, and emphasizing the need for companies to make it convenient for consumers to complain) and John Goodman & Steve Newman, Six Steps to Integrating Complaint Data into QA Decisions, 36 Quality Progress, Issue 2 (Feb. 1, 2003) (stressing the importance of complaint data in helping to identify issues with products and the data's effectiveness in assisting companies with resource allocation decisions to address quality assurance issues).

         The disclosure of EWR consumer complaint information is likely to discourage companies from actively pursuing consumer complaints and to lead companies to limit their ability to receive consumer feedback. The fewer inputs that a company receives, the less reliable the information available to it and the less useful the data is to NHTSA to evaluate the field experience of a product. EWR consumer complaint data are particularly important to NHTSA in light of the fact that the agency commonly receives far fewer complaints than manufacturers, field report numbers are but a fraction of complaint numbers, and the warranty data are limited after warranties expire. The disclosure of consumer complaint data and attendant likely reduction in consumer complaint data would threaten the agency's ability to obtain robust complaint data.

         Consumer complaint data are competitively sensitive as well. The data would provide competitors with information on the performance of not only a particular vehicle but also of key, individual components. The EWR complaint data would provide information on product acceptance, perceived problems and vehicle and equipment systems that a manufacturer deems important. In view of their commercial value on sensitive performance and market issues, the disclosure of EWR consumer complaint data would cause substantial competitive harm to the manufacturer. Moreover, as with warranty data, actual and potential consumers could make cross-company comparisons, which would further result in competitive harm.

         The record supports maintaining the confidentiality of consumer complaint information. For example, the Alliance noted the value of EWR data, including complaints, in revealing customer satisfaction and manufacturer cost information. PCLG's petition provides no factual rebuttal.

         While PCLG accurately noted that NHTSA releases consumer complaint data in individual investigations, these limited disclosures with respect to specific models and model years are not comparable to the wholesale, industry-wide information comprising EWR data. As such, disclosing EWR complaint data would provide a competitor with commercially valuable information without making the necessary investment in research ordinarily required if the information were not made readily available. This point was echoed by a number of manufacturers, including the Juvenile Products Manufacturers Association (``JPMA'') (complaints reveal operational marketing strengths and weaknesses to expose company vulnerabilities) and AIAM (wholesale complaint disclosure eliminates the risks associated with producing and marketing a particular technology).

         In general, PCLG's petition mentions consumer complaints along with warranty claims, without a particular discussion of consumer complaints. See PCLG Petition for Reconsideration at 7, 9-10. Accordingly, in response, we refer the reader to the discussion pertaining to EWR warranty claims above. In addition, the following supplements the discussion above, with regard to consumer complaints.

         Like EWR warranty claim data, consumer complaints are not necessarily related to safety issues. Consumer complaints include expressions of dissatisfaction and claims of unsatisfactory performance of a product as well as assertions about an alleged defect. The problems may merely be alleged by an owner of a vehicle or may be real. The perceived problems addressed may involve performance that does not meet the expectations of the owner, but may not be significant. They may or may not be safety-related.

         NHTSA also balanced the importance of consumer complaints and the extent of the impairment to the government against the public interest in disclosure. The importance of complaints is well-established. The magnitude of the numbers of complaints is important to us, as in our screening we will look for trends based in part on relatively high numbers. We believe that, given manufacturers' substantial control over information collection, if the numbers of consumer complaints were disclosed to the public, it is likely that the numbers of consumer complaints would be reduced considerably and, as a consequence, our ability to detect potential safety problems would be substantially diminished.

         On the other hand, the public interest in disclosure of consumer complaints is limited. Standing alone, they simply indicate consumer dissatisfaction or perception of a potential or actual defect, by system or component. They do not identify the particular part or a problem. Based on complaint data alone, it is not possible to identify a safety defect in a particular product. Thus, the impairment prong balancing weighs in favor of nondisclosure of consumer complaint data.

         Further, as indicated in our discussion on EWR warranty data, the legal framework established by Worthington Compressors permits the consideration of possible consumer misuse of commercial information in determining the confidentiality of information under Exemption 4. In this instance, the record supports our view that consumer misuse of EWR complaint data is likely to occur. Comments from the Alliance (disclosure would facilitate misleading comparisons), AIAM (misleadingly high numbers might be due to differences in collection policies), JPMA (data have a great potential to mislead consumers) and others describe the manner in which these data are subject to misuse.

   6. Data Quality Act

      The RMA asserts that the Data Quality Act provides an independent basis to prohibit the disclosure of the EWR data the agency determined is not within the purview of Exemption 4. The RMA believes that the agency's release of EWR data would reasonably suggest to the public that the agency agrees with the data and would be relied on by the public as official NHTSA information. The RMA asserts the EWR information is subject to the Data Quality Act because it is factual data prepared by third parties, and in the RMA's opinion, not covered by any of the 12 exceptions contained in the DOT guidelines. The RMA also argued that the final rule does not meet the Data Quality Act's ``utility'' requirement and as written would not present manufacturers' data in an accurate, clear, complete and unbiased manner and in a proper context.

      We disagree. The early warning program is not subject to the requirements of the Data Quality Act because it falls within an express exemption. The OMB guidelines define the dissemination of information as agency initiated or sponsored

      [[Page 21424]]

      distribution of information to the public, but does not include responses to requests for agency records under the Freedom of Information Act, the Privacy Act, the Federal Advisory Committee Act or other similar law. (67 FR 8460). Thus, the Data Quality Act does not apply to data that the agency is required to disclose under FOIA but only to information that the agency discloses as part of an agency- initiated or sponsored dissemination of information.

      Consistent with OMB's guidance, the Department of Transportation developed a set of guidelines on information dissemination, which includes an exception for ``responses to requests under FOIA, Privacy Act, the Federal Advisory Committee Act or other similar laws.'' \22\ The information not covered by a class determination of confidentiality, or otherwise protected by a FOIA exemption, must be released under FOIA.

      \22\ DOT's Information Dissemination Quality Guidelines, at 12 (Effective Oct. 1, 2002). The DOT guidelines are available for public inspection at http://dms.dot.gov (click on the ``Data

      Quality'' link and then ``guidelines'').

      The process established by part 512 allows the agency to make available to the public information subject to FOIA by determining in advance which information is entitled to protection under a FOIA exemption. The FOIA provides the analytic foundation for the determination of which data will be publicly available and which will be protected from public disclosure. Accordingly, this information qualifies under the FOIA exception created by the OMB guidelines.\23\

      \23\ The FOIA mandates that the agency make broadly available information that has already been the subject of a FOIA request granted by the agency. An agency make available for public inspection and copying ``records * * * which have been released to any person [under FOIA] and which, because of the nature of their subject matter, the agency determines have become or are likely to become the subjet of subsequent requests for substantially the same records.'' 5 U.S.C. 552(a)(2)(D). In addition, under the the Electronic-FOIA Amendment of 1996, the information, if created after November 1, 1996, must be made available in an electronic format to the public. 5 U.S.C. 552(a)(2)(E).

3. Regulatory Analyses and Notices

   1. Executive Order 12866 and DOT Regulatory Policies and Procedures

      NHTSA has considered the impact of this rulemaking action under Executive Order 12866 and the Department of Transportation's regulatory policies and procedures. This rulemaking action is not significant under E.O. 12866, ``Regulatory Planning and Review'' or the Department's regulatory policies and procedures. There are no new significant burdens on information submitters or related costs that would require the development of a full cost/benefit evaluation. This rulemaking document will not change the impact of the final rule.

   2. Regulatory Flexibility Act

      We have considered the effects of this rulemaking action under the Regulatory Flexibility Act (5 U.S.C. 601 et seq.) This rule does not have a significant economic impact on a substantial number of small entities. Therefore, a regulatory flexibility analysis is not required for this action. This final rule imposes no additional obligations on the submitters of information to NHTSA beyond those otherwise required by the Vehicle Safety Act and the early warning reporting regulation with respect to the submissions of requests for confidentiality. This final rule addresses the agency's treatment of early warning reporting data and simplifies procedures for all submitters, including small entities, when submitting information to the agency. The rule protects from disclosure early warning reporting information found likely to cause competitive harm. It permits the disclosure of that early warning information determined neither to cause competitive harm if released nor to impair the ability of the government to obtain the information in the future.

   3. National Environmental Policy Act

      NHTSA has analyzed this rule for the purposes of the National Environmental Policy Act and determined that it does not have any significant impact on the quality of the human environment.

   4. Executive Order 13132 (Federalism)

      The agency has analyzed this rulemaking action in accordance with the principles and criteria set forth in Executive Order 13132 and has determined that it does not have sufficient federalism implications to warrant consultation with State and local officials or the preparation of a federalism summary impact statement. The rule has no substantial effects on the States, or on the current Federalism-State relationship, or on the current distribution of power and responsibilities among the various local officials.

   5. Unfunded Mandate Reform Act

      The Unfunded Mandate Reform Act of 1995 requires agencies to prepare a written assessment of the costs, benefits and other effects of proposed or final rules that include a Federal mandate likely to result in the expenditure by State, local or tribal governments, in the aggregate, or by the private sector, of more than $100 million annually (adjusted for inflation with base year of 1995). This rule will not result in the expenditure by State, local or tribal governments, in the aggregate, or by the private sector, of more than $100 million annually.

   6. Executive Order 12778 (Civil Justice Reform)

      This rule will not have any preemptive or retroactive effect. This action meets applicable standards in sections 3(a) and 3(b)(2) of Executive Order 12988, Civil Justice Reform, to minimize litigation, eliminate ambiguity, and reduce burden. The rule does not require submission of a petition for reconsideration or other administrative proceedings before parties may file suit in court.

   7. Paperwork Reduction Act

      The existing requirements of part 512 are considered to be information collection requirements as that term is defined by the Office of Budget and Management (OMB) in 5 CFR Part 1320. Accordingly, the existing Part 512 regulation was submitted to and approved by OMB pursuant to the Paperwork Reduction Act (44 U.S.C. 3501 et seq.). These requirements were approved through February 28, 2005. This final rule does not revise the existing currently approved information collection under Part 512.

   8. Executive Order 13045

      Executive Order 13045 (62 FR 19885, April 23, 1997) applies to any rule that: (1) Is determined to be ``economically significant'' as defined under E.O. 12866, and (2) concerns an environmental, health or safety risk that NHTSA has reason to believe may have a disproportionate effect on children. This regulatory action does not meet either of these criteria.

1. Data Quality Act

   Discussion of the impact of this rule and the Data Quality Act are discussed in the analysis contained in the preamble above. For the reasons discussed in that section, any dissemination of information pursuant to this regulation will not be subject to the Data Quality Act.

   10. Regulation Identifier Number (RIN)

   The Department of Transportation assigns a regulation identifier number (RIN) to each regulatory action listed in the Unified Agenda of Federal Regulations. The Regulatory Information Service Center publishes the Unified Agenda in April and October of each year. You may use the RIN contained in

   [[Page 21425]]

   the heading at the beginning of this document to find this action in the Unified Agenda.

   List of Subjects in 49 CFR Part 512

   Administrative practice and procedure, Confidential business information, Freedom of information, Motor vehicle safety, Reporting and recordkeeping requirements.

   0 In consideration of the foregoing, the National Highway Traffic Safety Administration amends 49 CFR Chapter V, Code of Federal Regulations, by amending part 512 as set forth below.

   PART 512--CONFIDENTIAL BUSINESS INFORMATION

   0 1. The authority for Part 512 continues to read as follows:

   Authority: 49 U.S.C. 322; 5 U.S.C. 552; 49 U.S.C. 30166, 49 U.S.C. 30167; 49 U.S.C. 32307; 49 U.S.C. 32505; 49 U.S.C. 32708; 49 U.S.C. 32910; 49 U.S.C. 33116; delegation of authority at 49 CFR 1.50.

   0 2. Revise paragraph (c) of 49 CFR 512.21 to read as follows:

   Sec. 512.21 How is information submitted pursuant to this part treated once a confidentiality determination is made?

   * * * * *

   (c) Should the Chief Counsel, after considering a petition for reconsideration, decide that information is not entitled to confidential treatment, the agency may make the information available after twenty (20) working days after the submitter has received notice of that decision from the Chief Counsel unless the agency receives direction from a court not to release the information.

   0 3. Amend Appendix B to Part 512 by revising the first paragraph to read as follows:

   Appendix B to part 512--General Class Determinations

   The Chief Counsel has determined that the following types of information would presumptively be likely to result in substantial competitive harm if disclosed to the public:

   * * * * *

   0 4. Amend Appendix C to Part 512 by revising paragraphs (a)(2) and (a)(3), by adding a new paragraph (a)(4), and by adding a new paragraph (c) to read as follows:

   Appendix C to Part 512--Early Warning Reporting Class Determinations

   (a) * * *

   (1) * * *

   (2) Reports and data relating to field reports, including dealer reports and hard copy reports;

   (3) Reports and data relating to consumer complaints; and

   (4) Lists of common green identifiers. * * * * *

   (c) The Chief Counsel has determined that the disclosure of the last six (6) characters, when disclosed along with the first eleven (11) characters, of vehicle identification numbers reported in information on incidents involving death or injury pursuant to the reporting of early warning information requirements of 49 CFR part 579 will constitute a clearly unwarranted invasion of personal privacy within the meaning of 5 U.S.C. 552(b)(6).

   Issued on: April 16, 2004. Jeffrey W. Runge, Administrator.

   [FR Doc. 04-9005 Filed 4-20-04; 8:45 am]

   BILLING CODE 4910-59-P