NHTSA Safety Defect Investigations

NHTSA and the Safety Act

The National Highway Traffic Safety Administration (NHTSA), U.S. Department of Transportation, is the federal agency responsible for issuing and enforcing safety standards concerning the manufacture of motor vehicles and motor vehicle equipment, and for investigating alleged defects in those vehicles and equipment.[2] The basic purpose of the Safety Act that NHTSA administers is “to reduce traffic accidents and deaths and injuries resulting from traffic accidents.”[3] The need for a federal law on safety defects is noted in the Safety Act’s legislative history: “Deficiencies in past industry practices relating to notification and curing of manufacturing defects necessitate the imposition of mandatory procedures to insure such notification of purchasers and correction of all safety-related defects.”[4]

Today I will talk to you about the Safety Act requirements with respect to the recall of defective products, how the case law defines safety-related defect, and how NHTSA investigates to find those defects. I’ll highlight how the TREAD Act[5] passed last year will affect defect investigations. Finally, I’ll offer a few suggestions about how you can influence NHTSA’s defects investigations, and things you want to avoid doing.

Duty to Recall Vehicles and Equipment Containing a Safety-Related Defect

The Safety Act requires a manufacturer who determines that a safety-related defect exists in its vehicles or equipment to notify purchasers of safety-related defects and to remedy the defect, which is commonly known as a recall.[6] However, the public is not left solely at the mercy of the manufacturer to make a safety defect determination. The statute also gives NHTSA the authority to investigate defects[7], and to require manufacturers to recall when NHTSA finds that the vehicle or equipment contains a safety-related defect.[8] In other words, if the manufacturer does not “voluntarily” recall vehicles or equipment containing a safety-related defect, NHTSA may order it to do so, and may also seek to penalize the manufacturer for failing to fulfill its recall responsibilities. Given this duty of manufacturers to recall products containing a safety-related defect, and NHTSA’s authority to investigate and order recalls of safety-related defects, it is important to understand what is meant by the term “safety-related defect,” and how one is identified.

What is a Safety-Related Defect Under the Safety Act?

The concept of a safety-related defect under the Safety Act can be broken down into two steps: (1) is there a defect, and, if so, (2) does it relate to motor vehicle safety?

The Safety Act provides two relevant definitions:

1. “defect” includes any defect in performance, construction, a component or material of a motor vehicle or motor vehicle equipment. [49 U.S.C. § 30102(a)(2)]
2. “motor vehicle safety” means the performance of a motor vehicle or motor vehicle equipment in a way that protects the public against unreasonable risk of accidents occurring because of the design, construction, or performance of a motor vehicle, and against unreasonable risk of death or injury in an accident, and includes nonoperational safety of a motor vehicle. [49 U.S.C. § 30102(a)(8); emphasis added]

“Unreasonable risk” is not defined in the Safety Act. Moreover, the definition of “defect” is of limited help. For example, it does not define a defect in performance. Fortunately, court decisions in defect cases brought by NHTSA and discussed below have helped define the scope of those terms.

Court Decisions Construing “Safety-Related Defect” Under the Safety Act

Wheels Case Construes “Defect” Under the Safety Act

The Wheels decision[9] is the seminal case on the definition of the term “defect” in the Safety Act. That case involved alleged broken wheels in GM pickup trucks. GM acknowledged that the wheels (manufactured by its supplier, Kelsey-Hayes) were indeed breaking, but contended that NHTSA had not shown that the failures did not result from owner misuse. NHTSA relied primarily on the large number of failures to prove a “defect in performance.” The agency obtained from GM 2361 unverified reports of wheel failures, and from a random sample of them, obtained 160 owner affidavits, and introduced statistical evidence extrapolating this sample to the larger population of reports. The agency filed a motion for summary judgment based on these affidavits, arguing that the large number of failures proved, as a matter of law, the existence of a performance defect. The district court agreed, and granted the summary judgment motion. GM then appealed.

The U.S. Court of Appeals for the D.C. Circuit upheld NHTSA’s view that a defect in performance in a motor vehicle component could be proved by showing a large number of failures in that component.[10] The court held that “a vehicle or component “contains a defect’ if it is subject to a significant number of failures in normal operation, including failures either occurring during specified use or resulting from owner abuse (including inadequate maintenance) that is reasonably foreseeable.”[11] “Significant number of failures” means a number of failures that is “non-de minimus:”

The question whether a “significant” number of failures have taken place must be answered in terms of the facts and circumstances of each particular case. Relevant considerations include the failure rate of the component in question, failure rates of comparable components, and the importance of the component to the safe operation of the vehicle. The number of failures need not be and normally will not be a substantial percentage of the total number of components produced.[12]

The court in Wheels found that it was not a sufficient defense for a manufacturer to assert that owners exceed weight limitations or fail to properly inflate their tires to manufacturers’ specifications, provided that the owner abuse was not so gross as to be unforeseeable, since “owners do not always pay close attention to the weight of their cargo, and are not always scrupulous to make sure that tires are inflated to specified pressures, and loads carefully balanced front-to-back and side-to-side.”[13]

The court held that a “commonsense” approach requires manufacturers to build in some margin of safety to protect against failure during day-to-day operation, although not against gross owner abuse. As an example of a commonsense approach, the court indicated that it would not find defective a wheel that collapsed under a load four times that specified by the manufacturer, but that twice the load specified could be deemed foreseeable.[14]

In analyzing the Safety Act’s language, the court stated that the statutory “provisions indicate that a determination of “defect’ does not require any predicate of a finding identifying engineering, metallurgical, or manufacturing failures. A determination of “defect” may be based exclusively on the performance of the vehicle or component.”[15]

Moreover, The Wheels court noted the importance of real life performance:

Congress was concerned with the day-to-day performance of motor vehicles in the myriad of conditions of use experienced by the public, not the test data compiled by professional drivers on the manufacturer’s proving grounds or performance specifications under laboratory conditions.[16]

The court of appeals in Wheels thus clarified the meaning of “defect” in the Safety Act. It did not, however, clarify the issue of “unreasonable risk,” since it did not reach that issue.[17]

Pitman Arms Case Construes “Unreasonable Risk” Under the Safety Act 

The Pitman Arms decision is the seminal case on safety-relatedness.[18] In that case, NHTSA contended that the failure of a critical safety system (steering) creates an unreasonable risk of accident occurrence. At trial, the Government showed that a large number of failures had occurred. GM contended that failures cannot occur at high speeds, and do not pose an “unreasonable risk” at low speeds. It argued that the vehicles, 1959 and 1960 Cadillacs, had traveled 24 billion miles by the 1974 trial, that 96% of their service life was over, with an average of only 3 years of service left on 40,000 vehicles still on the road out of 284,000 that had been manufactured, and that with that service history, there were only 19 known accidents, and no deaths or injuries.

The trial court found that a large number of replacement parts sales, some 26,000 or 9.3% of the original vehicle population, was a strong indication of a large number of failures, but held that the Government failed to prove that these failures imposed an unreasonable risk of accidents, injuries, or deaths, since the failures occurred while the moving at low speeds (less than 10 mph). On appeal, the Government argued that low speed failures do present an unreasonable risk, noting accident statistics indicating that a significant portion of all accidents, injuries, and deaths do occur at low speeds. The court of appeals reversed the trial court, stating:

The evidence is uncontradicted that General motors sold six times as many pitman arms for the 1959-60 Cadillac models as for adjacent years; that the steering pitman arm failures have occurred while these models were being driven; and that when the steering pitman arm fails, the driver loses control of the car. We hold that, under the statute, these uncontradicted facts demonstrate an “unreasonable risk of accidents” stemming from the defect.[19]

Thus, while the Wheels court talked about a “non-de minimus” number of failures as evidencing a defect, the Pitman Arms court helped to quantify that formulation, as well as to clarify when the defect is safety-related. First, where a component fails at six times the rate of its peers, the component contains a “defect” within the meaning of the Safety Act. This is not to say that a failure rate of less than six times a peer group (e.g., four or five times the peer group) is automatically acceptable (i.e., non-defective), but the multiple of six is ordinarily deemed to be well within the ambit of what is unacceptable.

Second, if the failure of the component can result in loss of control of the vehicle, then the defect is deemed to be safety-related within the meaning of the Act. This is not to say that there cannot be other manifestations of safety relatedness, but the Pitman Arms decision does mean that a component failure that can result in loss of vehicle control is ordinarily deemed to be safety-related.

Subsequent Cases Follow Wheels and Pitman Arms

In subsequent litigation, courts accepted the holdings in the “Wheels” and Pitman Arm decisions, in a variety of fact patterns where there was “thin” or no evidence of actual fatalities or serious injuries resulting from the defect at issue. For example, in the Seatbacks case[20], the court found that a defect is safety-related where the driver’s seatback suddenly fails, inasmuch as the driver may fall rearward, and may loose contact with the brake and accelerator or experience impaired ability to break or steer, and may loose vision of the road or contact with steering wheels. There were no known fatalities attributable to the defect, but the court concluded that “the causes of many fatal accidents is [sic] never known, for dead men tell no tales.”[21]

In the Carburetors case,[22] the court of appeals noted that just as failures that cause loss of control constitute an unreasonable risk, so too do failures that result in engine fires, which may occur on thoroughfares where pulling over and standing outside the car is difficult and dangerous.[23] Where a defect has been established, and “where this defect results in hazards as potentially dangerous as a sudden engine fire, then the defect must be viewed as one “related to motor vehicle safety.”[24] And in the Wipers case,[25] the court found that since windshield wiper failure results in severe degradation of visual performance, it is an unreasonable risk to operate a vehicle in rain without windshield wipers.

NHTSA’s litigators developed the “per se” legal theory in attempting to establish that defects are safety-related. The theory is that if a defect causes the failure of critical equipment, components affecting vehicle control, then it is necessarily safety-related. Thus, for example, if a defect causes the driver to loose control of steering or braking, it is a safety-related defect per se. To some extent, the courts in Pitman Arms, Seatbacks, Carburetors, and Wipers cases seemed to be persuaded by the per se theory, although it was not essential to their holdings.[26] But it became accepted doctrine in NHTSA that if a critical component or system designed to control the vehicle, avoid accidents, or mitigate serious injury in the event of a crash, is defective, then that defect is safety-related. For example, it would not matter if no one has been injured (yet) when headlamps or seat belts fail in use.[27] If such components contain defects that cause them to fail, the defects are per se safety-related defects.

The X-Cars case[28] is the only defect litigation under the Safety Act in which the Government did not prevail. In that case, the Government relied upon consumer complaints to prove the existence of the alleged defect of premature rear brake lock-up, and did not introduce evidence that any particular component or design caused the alleged defect. The court of appeals held that the Government did not prove that the vehicle itself was defective, i.e., that failure had occurred, since the symptoms of the alleged defect reported in the consumer complaints (significant yaw while braking) are not always an indication that the braking system is performing improperly, and there were no broken parts evidencing a failure, as there had been in the earlier Safety Act cases. However, the court of appeals characterized the X-Cars case as a “highly fact-specific” (where there were no broken or failed parts evidencing the alleged defect), and the court not only did not overturn Wheels or Pitman Arms, but actually noted “happily” that those two cases had explained the meaning of the Safety Act’s term “defect.”[29]

Why Has There Been No Further Litigation?

What is remarkable is not that the Government lost only one fact-specific case, but that there has been no further litigation contesting the agency’s determination that a safety-related defects exists. Why is that? Some have argued that the agency has become atrophied, and is not enforcing the Safety Act. That argument is usually accompanied by a suggestion that NHTSA is too close to the industry it is supposed to regulate. But that ignores the fact that record numbers of vehicles are being recalled, frequently as a result of ODI investigations.

My own interpretation is that the basic law was settled in a manner very favorable to the Government in Wheels and Pitman Arms, and the manufacturers have grudgingly accepted those cases as the law. Consequently, when ODI, in concert with agency enforcement attorneys, believes that a vehicle or equipment contains a safety-related defect, and the manufacturer cannot disabuse the agency staff of that belief, the manufacturer will virtually always recall the product, rather than go through formal proceedings leading to a NHTSA recall order, let alone litigation to enforce that order. That does not mean that ODI always reaches the right result, or that what is “right” is always clear and always prevails, and that all vehicles with safety-related defects are recalled. It does not even mean that manufacturers always play honestly by the rules. They don’t always. But it does mean that manufacturers have been unwilling, when push comes to shove, to contest a NHTSA safety-defect determination made in light of the Wheels and Pitman Arms precedents. In fact, what has been happening since the Wheels and Pitman Arms decisions is that the manufacturers have acquiesced to the agency’s informal position by the end of a defect investigation, before a formal agency defect determination is made. With hardly any exceptions, when NHTSA and a manufacturer come eyeball-to-eyeball on a safety defect controversy, the manufacturer blinks first.

NHTSA’s Investigations of Safety Defects

The Office of Defects Investigation

The Safety Act gives NHTSA the authority and responsibility to, among other things, investigate defects in motor vehicles and motor vehicle equipment. I have heard reports of defendant manufacturers in product liability cases trying to portray the federal government in Washington, D.C. as a massive professional staff that thoroughly reviews, tests, and even “approves” vehicles before they are allowed to be marketed. This hardly comports with reality. NHTSA is a relatively small agency, with only about 600 employees, and many other important responsibilities. Only approximately 100 of the agency’s workforce are assigned to the enforcement area, about half of which are assigned to the Office of Defects Investigation (ODI). (The other half of the personnel in the enforcement area work in areas other than defect investigations, such as standards compliance and odometer fraud.)

Based on my 25 years as counsel to ODI, I can tell you that there are hard working, competent, and dedicated professionals there, but the office is generally overworked and understaffed. ODI is responsible for investigating safety-related defects in the nation’s fleet of over 220 million motor vehicles,[30] and hundreds of millions of items of motor vehicle equipment, including tires and child safety seats. Of ODI’s staff of approximately 50, only some 15-20 have been engineers or investigators who directly run safety defect investigations. The remainder are involved in other functions such as information management, screening of complaints, and central management. On average, ODI receives between 40,000 and 50,000 complaints annually, which come to the agency through various sources, including NHTSA’s toll-free consumer Hotline, other phone calls, NHTSA’s web page, E-mails, and letters. Although the number of vehicles on the nation’s highways increased, and they became more complex, the ODI’s working capital budget (in non-inflation adjusted dollars) and staff remained relatively constant throughout my tenure at NHTSA. In a typical year, ODI opens 80-100 defect investigations, of which more than half result in recalls.

Only since the passage of the TREAD Act, which was signed into law on November 1, 2000, has ODI been given significantly greater resources, which is now resulting in the hiring of approximately 18 new personnel, as well as funding for additional contractor support.[31] Eventually, those additional resources will help, but, ironically, in the short run they probably have been more of an impediment. Precious resources have been used in the past several months on administrative reorganizations, job descriptions, interviewing and hiring, and the logistics of finding additional office space, moving, and other attendant disruptions. The new staff now being hired has to be trained by existing staff. So the transition to a larger and reorganized staff has probably impeded ODI’s effectiveness in the short run, but there is no doubt in my mind that in the long run the defect investigations program will be more effective as a result of the added resources.

Now let me talk about the defect investigation process itself.

ODI’s Screening Process

Screening is the process of reviewing and analyzing the vast amount of information available to ODI, and making recommendations as to which matters should be opened as defect investigations. Since the agency would not ordinarily investigate or order the recall of a defect specific to a single unit, it looks for trends in larger populations. ODI’s Defects and Recall Information Analysis Division[32] performs this process by reviewing incoming reports of alleged defects, and identifying problems that may indicate a safety-related trend.

The data reviewed in the screening process include vehicle owner questionnaires (VOQs)[33] and field reports received from Congressional and owner correspondence, as well as referrals from other government entities (including other federal agencies, state agencies, and Transport Canada)) and from non-governmental organizations such as consumer and public interest advocacy groups, fleet owners, insurance companies, and media. In addition, ODI looks at technical service bulletins (TSBs) sent by manufacturers to their dealers. A NHTSA regulation[34] requires each manufacturer to furnish to NHTSA on a monthly basis a copy of all notices, bulletins, and other communications sent to more than one manufacturer, distributor, dealer, lessor, lessee, or purchaser, regarding any defect, regardless of whether the manufacturer believes that the defect is safety-related. ODI actually secures a large number of recalls by reviewing those TBS’s, and prodding the manufacturer to formalize its advice or “hidden warranties” into a statutory recall campaign. Finally, the screening process may include a review of NHTSA data bases (e.g., NASS, FARS) and past defect investigations.

The Initial Evaluation (IE) is the first formal step taken with regard to information concerning a potential safety-related defect. However, an IE is not a formal investigation so much as an internal record keeping designation for issue tracking. At this early stage, a low number of consumer complaints may be sufficient to justify initiating an IE. An IE may be opened by the Division Chief, upon the recommendation of a staff engineer/investigator. The IE may lead to a referral to one of ODI investigative divisions with a recommendation to open a defect investigation, which requires the approval of the ODI Director.

Early Warning Reporting Requirements: Future Donnybrooks

In the future, the screening process will undoubtedly include “early warning” data that manufacturers will be required to furnish to NHTSA. Provisions of the TREAD Act amendments to the Safety Act[35] require NHTSA to issue a regulation no later than June 30, 2002, that will require manufacturers to provide information related to (a) warranty and claims data concerning claims for deaths, serious injuries, and property damage from alleged defects; (b) customer satisfaction campaigns (“hidden recalls”), consumer advisories, recalls, and other activity involving the repair or replacement of vehicles or equipment; (c) other data that may assist in identification of defects, and (d) incidents causing fatalities or serious injuries allegedly caused by possible defects in vehicles or equipment in the United States, or in identical or substantially similar vehicles or equipment in a foreign country.[36]

While the agency is not necessarily required to disclose all of the information to be submitted, and manufacturers will be able to claim confidentiality for certain information, as they now do[37], presumably defendant manufacturers may be forced to disclose their “early warning” submissions to NHTSA in product liability cases, albeit under protective orders. This may prove to be a treasure trove of information to plaintiffs, and manufacturers are deeply concerned about that. I suspect that the problem for ODI will be not so much obtaining the information as being overwhelmed by huge “document dumps.” The difficulty will be separating the wheat from the chaff. While the industry will be in the uncomfortable position of gathering and reporting this information, ODI will be in the unenviable position of trying to sort it all out. And while not ODI’s concern, the General Law Division of the Chief Counsel’s Office will probably be swamped with confidentiality claims and FOIA requests. I can imagine NHTSA having to defend both FOIA suits (brought by public interest groups and possibly ATLA members seeking disclosure of early warning data) and so-called “reverse FOIA” suits (brought by manufacturers seeking to prevent disclosure). To the extent that the early warning data is disclosed, I think that plaintiffs attorneys and public interest groups will want to provide their own evaluation of that data to NHTSA. Stay tuned.

Formal Defect Investigations

ODI conducts defect investigations in two phases: the Preliminary Evaluation (PE) and the Engineering Analysis (EA). In the PE phase, ODI typically sends only one information request (IR) to the investigatee-manufacturer, and typically does not send IRs to any other parties. ODI does not typically conduct testing or inspect vehicles during the PE phase. An agency attorney is ordinarily not involved with an investigation at the PE phase. Most PEs are not upgraded to EAs, and are resolved within a few months. A PE may be closed because further investigation does not appear warranted, or because the manufacturer has decided to conduct a recall. When a PE or EA is closed where there has been no recall, it is typically justified on the basis that no defect trend has been identified, and that no further expenditure of resources appears to be warranted. However, contrary to some representations of the manufacturers and their attorneys, ODI does not make an affirmative finding of “no defect.” The ODI Director participates in decisions to open and close a PE, but agency counsel and higher management typically are not actively involved at the PE phase, unless there are unusual circumstances.

In the EA phase, ODI devotes more resources to the investigation, and develops a more detailed and complete analysis of the character and scope of the alleged defect. An IR, more detailed than that sent in the PE phase, is sent to the manufacturer, and after reviewing the response thereto, ODI may send one or more further IRs to the manufacturer. ODI may also send IRs to “peer manufacturers” and/or other companies. The agency may also conduct testing and vehicle (or motor vehicle equipment) inspections, either through NHTSA’s Vehicle Research Test Center in East Liberty, Ohio, or through contractors. ODI may also conduct surveys or other studies. ODI may meet directly with the manufacturer. Consumers who have reported problems may be contacted for further information. In addition to the ODI investigator, an agency attorney from the Office of the Chief Counsel’s Litigation and Enforcement Division is assigned to each EA. Experts from other offices in the agency may also be consulted. Crash data may be obtained from NHTSA’s National Center for Statistics and Analysis (e.g., FARS, NASS). Updates of complaints from the manufacturer’s and ODI’s complaint data bases are gathered. Thus, in the EA phase there is a much more intensive investigation than in the PE phase. Not only the ODI Director, but her boss, the Associate Administrator for Safety Assurance, are typically briefed on the progress of an EA, and if it is sufficiently sensitive, higher levels of management in the agency are also briefed and offered the opportunity to provide policy direction. An EA ordinarily takes about a year to complete, and may take longer, during which time an extensive investigative file is developed, updated, and closely analyzed.

During the course of a “mature” or largely completed EA, the manufacturer will be informally advised if ODI believes that a safety-related defect exists. If the manufacturer declines an informal request to recall, a multi-disciplinary review panel is convened, consisting of representatives of the Offices of the Administrator and the Chief Counsel, and other elements of the agency. At the panel meeting, the ODI staff investigator who has been assigned to the PE and EA will brief the panel about relevant facts. If the panel concurs, the ODI Director will advise the manufacturer of that fact and that a recall request letter (RRL) will be forthcoming within several days. This provides the manufacturer with a further opportunity to recall before a RRL is sent. If the investigation has gone this far, at this point the manufacturer will usually recall, rather than receive a formal RRL, which would be a matter of public record, and something that manufacturers do not like (and that ATLA members love). Virtually no cases go further, but if the manufacturer were to refuse to recall, the Associate Administrator can issue an initial decision pursuant to the Safety Act[38] that the vehicles or equipment at issue contains a safety-related defect. A final decision is reserved to the Administrator, but in practice, neither a RRL nor an initial decision is issued without concurrence from the Administrator’s office.

Manufacturers knows that when a panel meeting is scheduled concerning their products, that is their penultimate opportunity to avoid public controversy, and that when the panel concurs in the ODI recommendation and a RRL is about to be sent, that is their last chance to recall without a formal recall request from NHTSA on the record. That “threat” is virtually always sufficient to trigger a recall determination by the manufacturer, but if not, the agency can issue a statutory initial defect determination by NHTSA (accompanied by a press release), hold a public hearing, and then issue a final defect determination and recall order. In today’s media-hyped environment, no manufacturer wants to undergo such an ordeal. The continuing adverse publicity would be devastating. Challenging the agency’s recall order in court would bring more adverse media attention. General Motors’ experience with the X-Cars litigation could be described as winning the battle and loosing the war, as the publicity negatively impacted sales of the vehicles at issue. There is no market share to be gained in fighting NHTSA.

How to Properly Influence a NHTSA Investigation

Since the statutory defect determinations are virtually never made any more, the key time to influence the agency is during the ODI investigation. The Wheels and Pitman Arms cases provide a framework for the kinds of evidence that ODI seeks and evaluates in the course of a defect investigation. The more accident and injury reports that ODI has, the more likely it will be to pursue the investigation. Expert reports can also be helpful to ODI in establishing the existence of a defect. The agency traditionally has been receptive to information supplied by the plaintiffs’ bar, and lately has affirmatively sought such information in some tire investigations. There is nothing unethical or improper about providing such data to NHTSA, assuming that you are properly in possession of the information, and it is not covered by a protective order. You are providing a public service in reporting information to a safety agency.

The ODI investigator will appreciate that assistance. On the other hand, he or she will not appreciate subpoenas, or threats to subpoena their files or take their depositions. Simply put, information sharing with NHTSA is not a two-way street in a pending investigation. You can obtain a certified copy of a public file, but it will not include internal NHTSA evaluations, or information submitted by the manufacturer for which a confidentiality claim is pending or has been granted. The ODI staff are human beings, and the quickest way to make them unsympathetic to your client’s position is to criticize their professionalism, make personal attacks, or make unreasonable demands on their time. You should not expect to be successful in securing the testimony of agency staff, as the agency effectively resists having its defect investigators testify in civil product liability cases.

I would also recommend that you think twice before formally petitioning the agency to open an investigation. If you do not have good reason, perhaps based on informal communications, to believe that the agency would grant such a petition, you run the very real risk that the agency will deny the petition. The agency is required to grant or deny a petition within 120 days,[39] and given the limited ODI resources, a high percentage of defect petitions are denied. Your adversary will then point to that denial as evidence of the lack of a defect. Finally, while I know to never say “never,” I would almost never recommend petitioning the agency to reopen an investigation that it has previously closed. In theory such a petition may be entitled to consideration on its merits,[40] but in practice there is a very strong tendency to affirm previous decisions, particularly where the staff member responsible for that previous decision is still with the agency. Even if the previous closure or petition denial was a “close” question, the reconsideration will probably not be, since the agency does not like to reverse itself.

In closing, I would like to say that the NHTSA mission of protecting the public from safety-related defects is a noble one, and it has benefited from the support given by ATLA members over the years. Motor vehicles are increasingly more complex, and it is more difficult than ever to investigate alleged defects. With your help, NHTSA will be up to the challenge. If you understand how the agency conducts a defect investigation, and what kinds of information it needs to find a safety-related defect, you can be in a position to influence the outcome.