Don't TREAD On Me:
NHTSA has announced that TREAD Act reports will be presumed to be confidential. If, as some expect, NHTSA won't be able to analyze the flood of data it's about to receive, and it won't release that data to the public, what's the benefit of compliance beyond penalty avoidance?
What will the U.S. government do with all the warranty data it is now gathering from automotive manufacturers? So far, the National Highway Traffic Safety Administration has said the warranty data filed with its agency will be presumed to be confidential, but many doubt whether this position will survive an expected court challenge. Many also doubt whether NHTSA will be able to do a better job than the manufacturers already do when it comes time to analyze the data.
The warranty data now filed by automakers and other manufacturers with the U.S. Securities and Exchange Commission wasn't actually ever formally requested by the government, though it was implied by the requirements of Section 401 of the Sarbanes-Oxley Act of 2002. Rather, the SEC left the formulation of the rules for warranty data submission up to private groups such as the Financial Accounting Standards Board, which made them part of a much broader set of rules concerning the disclosure of corporate guarantees.
The Transportation Recall Enhancement, Accountability and Documentation Act, better known by its acronym as the TREAD Act, was passed by Congress following their exasperation over the years of delays that followed the discovery of a pattern in the tire failure data of Firestone and Ford. Had the manufacturers disclosed the data faster and initiated the recall sooner, Congress said, lives could have been saved.
Safety or Compliance?
NHTSA was directed to craft rules for the timely and comprehensive reporting by auto manufacturers of the product defects and failures that come to their attention and the deaths and injuries that result. Observers say the agency grossly underestimated the cost of TREAD Act compliance, not only on the industry, but also upon itself. While the problem before TREAD was not enough information, the problem hereafter may be too much information, which overwhelms the agency, goes largely unexamined, remains unavailable to the public, and ends up filed away in some government archive. If the initial goal of TREAD was to save lives, it seems to have degenerated into a simple act of compliance in the face of threats: file these reports on time or we'll make your lives miserable.
In the Oct. 27 edition of Warranty Week, we took a close look at the way a group of small trailer manufacturers was able to gain an exemption for themselves from the more comprehensive reporting requirements of the TREAD Act. In the Nov. 3 edition, we examined the incredible spread between NHTSA's initial estimates of the cost of compliance and the actual spending totals now being reported by the auto industry analysts at AMR Research Inc.
Bias Alert
In this, the concluding part of the series, Warranty Week takes a look at the processability and confidentiality of the TREAD Act reports as the first deadlines approach. Regarding the processability, steady readers are well aware of how many manufacturers now include warranty data in their SEC Form 10-Q and 10-K statements, and how complex the quarterly analysis of this data can become. One can only wonder how NHTSA will make sense of the reports it receives, and how the agency could ever resolve the reams of summaries and aggregate data it receives down to an individual and actionable point of failure. Merely calculating a number for warranty claims as a percentage of a manufacturer's total product sales is a complex operation. No wonder the SEC performs no analysis upon the warranty data already in its vast archives.
Regarding the confidentiality of NHTSA's warranty data, readers should not be surprised to discover that Warranty Week is firmly opposed to the rules NHTSA published in the Federal Register on July 28, 2003 (Vol. 68, No. 144, Pages 44209-44232). However, we believe that NHTSA should release the data only to us, and not to any of the consumer watchdog groups that are now petitioning the agency to reconsider its decision. We would relish the chance of cross-referencing NHTSA's data with the SEC's data to resolve warranty costs down to the vehicle or the component level, and to announce once and for all which manufacturer's vehicles incur the highest and lowest warranty costs. We want to tell you down to the penny how much GM spent in the third quarter on warranty claims per vehicle for the Buick Regal with a 3.8 liter V6 engine, leather seats, and aluminum wheels. And we want to compare that to the claims rates seen in similar vehicles made by Infiniti, Lexus, and Acura.
Alas, we'll probably never get that chance. First of all, comparative analysis of publicly-available warranty claims data has only been possible for the past six months or so, and even then only in net dollar terms based on the SEC data. Some say it won't be long before manufacturers themselves begin using this data in their own advertisements, given how some already use the mere existence of lengthy warranties to their own advantage ("America's Best Warranty," etc.). This is something of a nightmare scenario for some manufacturers, who will find their own somewhat accommodative warranty policies turned against them, if their willingness to pay borderline claims is miscast as a sign of high rates of failure for their products.
Imagine for a second if the quality and reliability surveys of Consumer Reports or J.D. Power and Associates were backed up by detailed government reports of failures per thousand vehicles accurate to three decimal places. Even with its vague crash test ratings, which rates vehicles like movies with one to five stars, NHTSA already suffers an incredible amount of criticism. Imagine if the NHTSA database becomes the source of advertising claims such as "America's Best Vehicle."
Second of all, NHTSA has already changed its TREAD Act rules several times, and if it did so again once the reporting process has begun the agency might be accused of making it up as it goes along. However, that's not to say that it is highly possible that NHTSA will discover after a quarter or two that some of the ideas expressed in numerous petitions for changes in the rules actually make sense. For instance, if NHTSA discovers that too many small companies were swept up by its arbitrary 500-vehicle-per-year line of demarcation between comprehensive and occasional reporting requirements, why not listen to those asking for an increase to 5,000 vehicles?
Approaching Deadlines
According to the NHTSA Web site, the one time historical reports are now due by Jan. 15, 2004 and they must cover the three-year period between July 1, 2000 and June 30, 2003. The first current report, covering the quarterly period between July 1 and Sept. 30, 2003, is now due by Dec. 1, 2003, which means it should be well on its way towards completion if it is to be filed on time.
Earlier this year, NHTSA's original timetable called for the first report to be filed by Aug. 31, covering the second quarter of 2003. That report was skipped, however, and the quarter it covered was folded into the historical reporting period. The Dec. 1 deadline for the third quarter report is to be followed by a Feb. 29, 2004 deadline for the fourth quarter report. In each of these cases, NHTSA is giving manufacturers roughly 60 days after the end of the quarter to file their reports. Once the electronic filing system has proven itself, reports will become due 30 days after the end of each quarter, which would mean regular deadlines each year on the last days of April, July, October, and January.
Stephen E. Selander, the senior counsel in the Detroit metro offices of Warner Norcross & Judd LLP, said he and others have petitioned NHTSA to make certain changes to the TREAD Act rules. A year ago, Selander petitioned NHTSA to increase the cutoff for vehicle manufacturers between Group 1 and Group 2 from 500 to 5,000 vehicles per year. He also thinks it would have been a good idea to eliminate altogether the reporting requirements for many of the parts and systems manufacturers in Group 2, because incidents involving these parts and components would already be reported by the final vehicle assemblers in Group 1.
"To me, a good portion of what they are requiring from component suppliers is basically going to be duplicative of what the vehicle manufacturers are supplying." He said he's quite willing to keep the tire manufacturers in Group 1, because after all, "Firestone is a tire manufacturer." He'd also leave the child restraint manufacturers in Group 1, conceding that there's no political support for any perceived lowering of safety standards for children. But he'd drop the reporting requirements for other component manufacturers.
Warranty Data Only?
Selander also suggested to NHTSA that it limit the one-time historical reporting to three years of just the warranty data, rather than requiring three years of everything. Those items are still in front of NHTSA, and NHTSA has not ruled, he said. Officially, NHTSA has said they remain under consideration, but has given no indication which way its decision will go. "I think those three things could be done pretty quickly to decrease the burden substantially, although it would decrease it now more in the ongoing area than in the setup," Selander said.
The AMR report suggests that NHTSA is unprepared for the amount of data it is likely to receive. "Consider that the industry processes $10 billion in claims annually and assume the average claim is $250," wrote AMR in its Sept. 26 report. "That's 40 million individual claims per year, and with the three years of historical data required, it's more than 120 million individual claims plus the related data."
The AMR report goes on to say that the analytical tools needed to sift through this data are not ready, and the tools to be used are under development and unproven. "With the amount of dollars that manufacturers invest in collecting the data and preparing it for submission, the tools that NHTSA will need to effectively monitor and detect patterns need to be cutting-edge as well," the authors wrote.
There is another aspect to this possible underestimation, namely that NHTSA itself also may face a compliance cost crisis if it truly believes that the act of reporting will be rare for Group 2 manufacturers, as was detailed in the Nov. 3 edition. Just the issuance of IDs to as many as 23,000 Group 2 manufacturers carries a processing cost much greater than zero, and just the archival of their occasional reports carries a storage cost much greater than zero. To minimize the workload, NHTSA has already told equipment manufacturers and low-volume vehicle manufacturers not to apply for an ID until they receive a reportable death claim or notice which alleges that a death was caused by a possible defect in the manufacturer's product.
Here Comes the Flood
Selander said that long before AMR reached its conclusions, NHTSA was warned that it could drown in a flood of its own making. Up until April of last year, in fact, Selander worked at General Motors, where one of his primary responsibilities was representing the automaker in matters before NHTSA. "Certainly, the manufacturers have told NHTSA it's going to get a huge amount of information and that it's going to be very difficult to figure out what to do with it," Selander said. "I think NHTSA is aware of that, and I think they are more confident in what their computer systems can do and what their data mining abilities are. And I know the AMR article questions whether they're really at that point. But I think that NHTSA plans at looking at certain aspects of that data more closely than other aspects."
Selander also questions how many more recalls will occur once NHTSA has more data and receives it faster. Based upon what he knows the Big Three already do when it comes to warranty claims analysis, Selander said he doesn't see where they have to gather any new information. "They may have to look at it a little differently, but the manufacturer has always had this kind of information, and there's nothing about this process that increases the information that the manufacturer gets. The manufacturer already has the legal obligation to do the recalls, and in terms of warranty-related things, certainly manufacturers already are looking for ways to reduce warranty cost," he said.
Civics Lessons for Prospects
James King, the vice president of TREAD Act solutions at Cypress Corp., said he sometimes finds himself giving civics lessons about the branches of government to the prospects he calls on the phone to talk about the impending deadlines.
Cypress sells a document management system, which it has decided to tailor for several rather narrow applications, such as TREAD Act reporting and the increased financial disclosure requirements that are part of the whole Sarbanes-Oxley Act, specifically the warranty tables mandated by FASB Interpretation No. 45.
While NHTSA is collecting merely numerical counts of specific types of failures, plaintiffs engaged in lawsuits will no doubt be aware that all of these numbers must tie back to more detailed stacks of documents. Some will want to see those stacks, and some courts will be inclined to grant them access to do so. In other words, for a manufacturer to file a TREAD report, they must first organize their documents. And by organizing their documents into neat piles, they are making it more likely that the courts will grant plaintiffs' specific requests for access to a certain pile. Before, the relevant data might have been stored in the basement chronologically, with a year's worth of data spread among hundreds of boxes containing thousands of fax pages. Now, everything to do with a specific component in a specific vehicle must at least be in the same box or in the same database. Otherwise, how could it have been counted and submitted to NHTSA?
King said he's calling major automotive manufacturers about TREAD, and is finding that they've confused it with FIN 45. The vice president of quality at one major glass manufacturer told him the company doesn't have to file TREAD reports because they're foreign-owned. So he had to explain how TREAD reports go to NHTSA, an agency of the Department of Transportation, while FIN 45 reports go to the SEC. NHTSA's rules cover all manufacturers whose vehicles are sold in the U.S., while the SEC's rules cover all companies whose shares of stock are sold in the U.S.
NHTSA vs. the Courts
Both NHTSA and the SEC ultimately report to the Executive Branch of government, which of course is headed by President George W. Bush. But all the product liability lawsuits take place in the court system, some of which are brought at the federal level and some of which proceed at the state or local level. Though the Executive Branch nominates the judges, once they take office their decisions are independent. State and local courts are even less dependent upon the Executive Branch. The bottom line is that the judges that preside over a civil court case report to neither the President nor his agencies, so while they are free to take advice from NHTSA or the SEC, they also are free to disregard the rulemaking activities of those agencies.
King noted that while NHTSA has so far decided to keep the TREAD data confidential, NHTSA is not part of the court system. If one day TREAD data is deemed to be potentially relevant in a court case, what's to stop a judge from granting access to that data? While he said NHTSA expects the data to remain confidential, he's not so sure that some judge somewhere won't one day require its disclosure during the discovery phase of a civil court case. And that comes down to a document management issue. If, for instance, a lawyer searching for a pattern in the data is granted access to all records pertaining to the sunroofs on a certain type of pickup truck made in 2002, a company that cannot be so specific in its data mining abilities is inviting the judge to grant the plaintiff broad access to all the company's sunroof data, or perhaps all the company's pickup truck data. In other words, companies that cannot provide specific documents are inviting the judge to allow the requesting attorney to go on a "fishing expedition."
King said he's heard that some of the major vehicle manufacturers are training their employees how to write factual reports that don't include damaging opinions or phrases such as "this could kill somebody." They're doing so because they fear that one day these documents might be read in court by a plaintiff to a jury. In other words, no matter what assurances they now get from NHTSA that their data will remain confidential, some companies are assuming that somewhere down the line this will change.
"Long term, I just can't see how NHTSA can hold off the legal profession from getting its hands on this information and it becoming public knowledge," King said. "It will hold up for a while, and NHTSA will protect them, but knowing the legal system ... those guys are tough!"
Petitioning for Reconsideration
Speaking of tough, NHTSA already has some tough critics who are prepared to fight for the public release of TREAD Act data if the agency doesn't change its rules. Joan Claybrook, president of Public Citizen and a former NHTSA official herself, said she is strongly in favor of the public release of the data filed by auto manufacturers as part of their TREAD Act Early Warning Reports. She noted that Public Citizen worked with several other consumer groups to file a petition for reconsideration of NHTSA's confidentiality rules.
Even though her term as head of NHTSA began and ended with the administration of President Jimmy Carter (1977 to 1981), Claybrook said she remembers how stubborn a bureaucracy can become once it's made a decision. " Normally, agencies don't change their mind," she told Warranty Week. "I don't know what they're going to do. I have absolutely no idea. It's not terribly likely that they will make major changes, because they have been working on it for two years. But I have no idea what their thought process is at this moment."
Clarence Ditlow, executive director of the Center for Auto Safety, which joined with Public Citizen in the petition for reconsideration, said he's also not sure what NHTSA will do next, but he knows what his group will do if NHTSA doesn't change its mind. "If they don't, we'll take them to court," he said.
By making TREAD Act reports public, Ditlow said, NHTSA would provide organizations such as his the statistical ammunition they would need to demand recalls and/or safety campaigns. "The main reason is to get the manufacturers to pay for their mistakes," he said. "The more information we have, the better the market will work."
Fear of Disclosure
Ditlow dismisses fears expressed by automakers that TREAD Act data could be misused by "ambulance chasing" attorneys, or by competitors looking to turn their claims, quality, and safety records into advertising copy.
"Those are just knee-jerk reactions," Ditlow said. "Remember, this is the industry that once said that airbags was going to cause the manufacturers to fold and go out of business. They once said that catalysts were going to cause the shutdown of assembly lines across the country."
Ditlow noted that NHTSA is collecting only numerical counts of the number of complaints, lawsuits, claims, and field reports, so it would be of limited use to an attorney in search of business. "You can't get names and addresses out of NHTSA. So you're not going to get clients," he said. Furthermore, while the data wouldn't be of much use to attorneys shopping for clients even if it were made public, it is likely to become available to attorneys who already have clients whether it's deemed to be confidential or not. "The people who will be able to get this information, whether it's confidential or not, are the product liability attorneys, because they have discovery powers," he said.
Ditlow also said he's seen some of the internal reports the major manufacturers generate regarding their own warranty claims, quality, and safety data. What they've been ordered to file with NHTSA is a small subset of what they already have. "I'm going to tell you -- they analyze their warranty data to a fare thee well. There was a whistleblower one time who sent us these boxloads of warranty printouts," he said. "I mean, if you wanted to know what component was failing more rapidly and more costly and which plant and which make and model, there it was." Therefore, why would attorneys even bother going after NHTSA's spreadsheets containing nothing but numerical summaries? They'd be better off going after the data the manufacturers keep confidential, even from NHTSA.
"All this law is intended to do is to give an agency which has proved to be singularly inept at finding defects in the past a better filter," Ditlow said. "The agency is requiring this information in a systematic format. Presumably, they're going to analyze it in a systematic format. But in an agency that does far more investigations into motorcycles than anything else on a per-unit basis, I have to wonder about their priorities."
FIN 45 Doesn't Impact TREAD
Neither Ditlow nor Claybrook thought the public availability of the SEC/FASB warranty data would have any bearing on their petition for public release of the TREAD Act data. "I don't know that it makes it moot," Claybrook said. "All the TREAD Act requires is the number. It doesn't require that each report has to be disclosed. I think they [NHTSA] are less interested in the dollar amounts, although that certainly can be an important piece of information. And in, fact, it seems to me [the dollar amounts] are more invasive from the company's point of view than just the number of reports."
Ditlow noted that the FIN 45 warranty tables that are now part of most manufacturers' SEC filings are, like the TREAD Act reports, merely summary numbers. While the TREAD reports contain a count of the total number of incidents, the FIN 45 tables provide just the net dollar amounts spent and accrued for warranty claims, after subtractions for supplier recoveries. Plus, the FIN 45 tables don't break those totals down by make or model, nor are they correlated to specific recalls or safety campaigns. Also, because the SEC's reporting requirements apply only to U.S.-based public companies, there are no warranty tables filed by DaimlerChrysler, Toyota, Honda, or others ultimately based in Europe, Japan, or Korea.
Ditlow also said the warranty spending and accrual numbers don't in and of themselves always correlate to quality or reliability statistics. A low spending number may be the result of a high number of claims that the manufacturer simply will not honor. "If a manufacturer takes a hard nose and says 'no, we're not going to do that. We're not going to pay this claim.' If one manufacturer turns down a $4,000 claim and another manufacturer just willingly pays it, you don't get a true picture of what the defects are in the vehicle," he said.
Back to Part Two |