Jayne O'Donnell, USA TODAY
- NHTSA says it tells consumers what they need to know for auto safety
- Automakers can request federal safety agency keep records confidential
Herman Ray Evans was killed when the tread separated on his 2001 Ford Explorer's tire, the vehicle rolled over into the median and he was ejected, according to the Daphne, Ala,. police crash report.
Evans was in one of at least 15 fatal tire-related crashes last year in Ford Explorers and Mercury Mountaineers, according to news reports. In Evans' crash, the police department did not find fault with the tiremaker or Ford Motor.
But from April 2002 through 2009, there were 375 similar deaths in mostly older-model Explorers and Mountaineers - nearly four times the number that led to the Ford/Firestone fiasco in 2000, according to an analysis of federal fatality data by the research firm Quality Control Systems. Still, neither the National Highway Traffic Safety Administration nor Ford Motor recalled the vehicles.
In a statement, Ford said it reviewed this issue with NHTSA and found the Explorer "has no unique tire issues." Ford says "numerous government safety agencies and independent organizations" have confirmed the Explorer's safety.
Although Ford is required to tell NHTSA about claims it receives about serious injuries and deaths in its vehicles - and NHTSA can investigate them - information about the probes and many of the injuries and deaths is only available to the public and news media through a Freedom of Information Act request. Even then, manufacturers can request the information they submit to the agency be kept confidential.
That's in sharp contrast to all the information posted on NHTSA's website about potential defects, injuries and deaths for the agency's official safety investigations. But informal investigations - where weeks or months can go by before potential problems are brought to the public's attention - are becoming more common. And that has ramifications for car buyers, who may not learn the vehicles they own or are considering buying have quietly raised safety concerns at NHTSA and among automakers.
• Hyundai recalled some of its Veracruz and Santa Fe SUVs in August 2011 because their air bags might not inflate in a crash. There was no public investigation or incidents, but in announcing the recall, Hyundai said it had been contacted by NHTSA about its nearly 8,000 warranty claims and 16 consumer complaints.
• Evenflo and NHTSA recalled more than 1 million Discovery child car seats in January 2008. But consumers didn't know the agency had been testing the seats for nearly a year because of concerns about how the seats fared in crash tests.
• And in November 2011, NHTSA announced it was opening a formal investigation of Chevrolet Volts - although it had been quietly testing Volts for six months after a car caught fire weeks after a crash test. The agency said it didn't consider that an investigation.
What these federal safety scenarios have in common is that they happened behind the scenes in what President Obama last week called "the most transparent administration in history" - even though questions about its transparency extend far beyond the halls of the U.S. Department of Transportation (DOT).
And they also involve an agency taken to task by Congress and DOT's Inspector General after the Ford/Firestone and the more recent Toyota sudden-acceleration cases. The issues: How NHTSA conducts its investigations of possible vehicle-related defects and how much information it publicly releases.
"It all goes back to the criticism for years: Where's the methodology? What are the thresholds? No one knows," says Sean Kane of Safety Research & Strategies. "It's a black art over there."
But NHTSA chief David Strickland disputes charges that his agency's methods are opaque and says if NHTSA isn't the most transparent agency in the federal government, it is certainly one of the most open.
"We definitely get information out to the consumer," he says.
The life cycle of a probe
Federal laws and regulations are both designed to prevent secrecy and written in ways that enable it.
Just what exactly signals the start of a formal public investigation falls into a gray area, both NHTSA critics and supporters agree. And that gives federal investigators considerable leeway when they want to work behind the scenes with the automakers they regulate.
The information that would be available to consumers and safety advocacy groups if early investigatory work were public includes testing, correspondence with manufacturers, injuries and deaths, says Allan Kam, a former NHTSA senior enforcement attorney.
On the other hand, delaying or deciding against starting a formal investigation can save automakers the bad press that comes from public probes - especially if they don't lead to a recall.
"Maybe the agency and the manufacturer know, but the public doesn't know and may want to provide input," Kam says.
NHTSA says the screening work it does to decide whether to open an investigation is sometimes confused with actually being an investigation. A public investigation is only opened when the agency has a reasonable belief a defect may have occurred, NHTSA says.
Strickland says NHTSA is able to do more pre-screening work these days, thanks to new data tools, "better technology and better exchanges with manufacturers." This non-public phase can last for months, depending on the complexity of the issue.
"Sometimes it's very fast; sometimes it lasts longer," he says.
But Kane, whose group does research for lawyers, engineering firms and the government, says he believes NHTSA carries this to an extreme so it can toil out of the harsh glare of public - and consumer advocate - scrutiny.
He calls NHTSA's work before it opened a formal probe of the Volt a "secret investigation" and questions how the agency could get a recall of the Evenflo car seat after doing the kind of testing usually considered part of a public investigation.
"If they can hide these kinds of investigations, there's more secrecy and it's less likely they're going to be accountable," Kane says.
To those in industry and government who suggest Kane pushes for public probes so the information is available to lawyers suing automakers, Kane notes, "The litigation isn't there unless people are killed or seriously injured."
The budget for NHTSA's Office of Defects Investigations nearly tripled in 2001, after all the news about Ford and Firestone injuries and deaths. Kane was instrumental in publicizing these issues after he uncovered what he calls a "secret recall" of tires by Firestone in Venezuela in 2000 that was then followed by a U.S. recall. In the past 20 years, the budget for the defects office has quadrupled.
Despite the increase and subsequent outcry over Toyota acceleration issues in 2009, the defects office has opened far fewer public investigations in each of the last three years than it had in nearly any of the previous 25 years.
"It seems logical to conclude that the reason there are fewer investigations is that they are doing more in the pre-investigation phase," says Kam, who worked in the defects office for 25 years until retiring in 2000.
Strickland says the agency shouldn't be judged by the number of public investigations it does: "The goal is to get recalls." In 2012, he notes, NHTSA defect investigations resulted in 138 vehicle recalls involving more than 9 million vehicles. That's the second highest number of recalls per investigation in the agency's history, he says.
The agency says it has gotten better at screening for defects, now that it gets what's known as "early warning reports" from automakers who have had to disclose data including warranty claims and injury and death reports to the agency since the post-Ford/Firestone law known as the TREAD Act. NHTSA also says it combs online car chat rooms and other sources for information.
NHTSA guidelines set timetables for how long each stage of an investigation should take, Kam says. That, he says, encourages the agency "to do more in the pre-investigatory phase because the clock doesn't start ticking until the investigation is opened."
Cars, trucks and child seats
These early-warning reports can lead NHTSA to launch a variety of little-known probes, including "Death Incident Inquiries," "Comprehensive Inquiries" and "Aggregate Inquiries" - all activities that resemble typical investigations but are done out of the public eye.
Advocacy group Public Citizen as well as Quality Control Systems separately sued the Transportation Department to get it to release the death and injury data it gets from automakers, and NHTSA agreed to release this information starting in 2008.
But Randy Whitfield, the statistician who heads Quality Control Systems, says he often has to make a FOIA request to get it.
Clarence Ditlow, who successfully led efforts in the late 1970s to get the informal early stage of defect investigations made public, believes information about NHTSA's death investigations eventually will become more public, too: "We'll get there," he says.
Kam, who now runs a highway-safety consulting firm that does work for companies or attorneys with matters before NHTSA, agrees the death reports should be readily available public information.
But if NHTSA were to do that, "Consumers wouldn't have any idea what to trust when it comes to the safety of their vehicles. There are fatalities every single hour of every single day," Strickland says.
Quietly testing car seats
NHTSA took Consumer Reports to task publicly for a January 2007 report that claimed some child seats were unsafe. The agency discovered the seats were tested at higher speeds than the magazine said.
The magazine retracted the story, but what NHTSA didn't make public was that it, too, began testing the Evenflo seat and asked the company for more information. That led to a January 2008 recall.
As for the months of investigating NHTSA did before it opened a public probe of the Volt, Strickland says he would have made the information public "if at any time we felt the public was at risk."
A year after congressional hearings in the Toyota matter, the DOT's IG report said NHTSA needed a "complete and transparent record system" that documents how it makes decisions in investigations.
The IG report noted even NHTSA's record keeping in public investigations was suspect: Of the 42 investigations the IG reviewed, the files didn't document meetings with manufacturers and third parties or describe its testing. The report concluded "process improvements" were needed to identify and address vehicle safety defects.
At the time of the report, NHTSA says it had already started taking action to address these issues.
Kam says he understands the reticence about public disclosure.
"If I was still at the agency, I wouldn't want people looking over my shoulder, scrutinizing the process while internal agency deliberations are still ongoing," he says. "Where you stand depends on where you sit."
Contributing: Meghan Hoyer