June 23, 2015
Inspector Agrees with SRS: NHTSA Ain’t Right
Today, we salute the good men and women of the Department of Transportation’s Office of Inspector General (OIG) for putting their official imprimatur on concerns that The Safety Record has been raising for years. The report, released to selected press on Friday, and to the rest of us slobs on Monday, was entitled: “Inadequate Data and Analysis Undermine NHTSA’s Efforts to Identify and Investigate Vehicle Safety Concerns.”
While the OIG said it in a much nicer way than we ever did – the 42-page report was a laundry list of issues we have long complained about: ODI’s lack of process, the prioritizing by chances of recall success rather than threat to safety; the agency’s penchant for shrouding everything in secrecy; the missing audits on manufacturer’s EWR audits and the lack of enforcement against the scofflaws.
The framework for the report was the GM ignition switch defect, which rapidly grew from a small recall in February 2014 with little information about how the automaker discovered the defect to a Greek tragedy in which the myriad institutional failing of twin protagonists NHTSA and GM took center stage at various hearings. After Deputy Administrator David Friedman testified in an April 2014 U.S. House hearing that an internal review of ODI’s policies and practices was underway, DOT Secretary Anthony Foxx ordered an external review.
EWR data was immediately flagged as a problem – only 24 component categories for 15,000 components, with no coding guidance for manufacturers, “resulting in inconsistent reporting and data that ODI investigative chiefs and vehicle safety advocates consider to be of little use.” Hmm. That sounds familiar.
In October 2012, The Safety Record published Further Tinkering with EWR Unlikely to Make it More Useful, decrying the agency’s proposal to make to add new EWR reporting categories relating to emerging technologies. We highlighted the observations of Alice and Randy Whitfield, who use EWR data extensively, have published their methodology in a peer-reviewed journal and have sued NHTSA successfully for access to EWR data.
"We suggest the implementation of a coding system for light vehicle deaths and injuries claims which links the category of the allegedly failing component with a separate code denoting the type of failure that is alleged. Such a system would take careful planning to propose and to put into practice. But it would be better to begin this planning now than to continue another nine years with an early warning system so lacking in necessary detail that NHTSA’s own analysts don’t rely on it for anything more than performance in a supporting role.”
Then the OIG rapped ODI for failing to have EWR audit procedures in place to verify that manufacturers submit complete and accurate early warning reporting data. We’ve been reporting the manufacturers are not submitting mandated EWR data since we discovered in February 2013 two manufacturers had neglected to file some death and injury claims. We had to sue the agency to find out what action, if any, that officials took to enforce the law. Hint: pretty much nothing. Here are some of the markers The Safety Record began laying down for the agency’s path forward nearly two years ago. You’re welcome:
October 2014 October 2014 Elective Warning Reports Redux
The OIG criticized NHTSA for failing to follow “standard statistical practices when analyzing early warning reporting data, such as establishing a base case for what statistical test results would look like in the absence of safety defects.” Actually, NHTSA has been turning its nose up to standard statistical practices in a whole lot of contexts, and we’ve been pointing this out for a while. You can read about it these stories:
July 2011 How NHTSA and NASA Gamed the Toyota Data
September 2011 NHTSAball: How the Agency Plays the Game
And non-profit, The Safety Institute, (founded by Safety Research & Strategies president, Sean Kane), established The Safety Vehicle Watch List for the sole purpose of using peer-reviewed analytic methods to bring a sane process to the task of identifying potential motor vehicle defects that merit further review, using EWR and the Fatality Analysis Reporting System.
The report also mentioned two of our longstanding bete noirs: lack of transparency:
And lack of process:
September 2012 The Pedal Error Error
We’ve got more links, docket comments, and lawsuits, but enough about our prescience. The report starred in the first panel of today’s U.S. Senate Commerce Committee hearing updating the members on the Takata recall efforts and NHTSA improvements. Sen. Claire McCaskill (D-MO) – ever so direct – distilled the picture painted by the OIG as “blatant, incompetent, mismanagement…if NHTSA isn’t clear about when to open an investigation we might as well shut it down.” NHTSA watchdogs Senators Edward Markey (D-MA) and Richard Blumenthal (D-CT) used the occasion to promote their NHTSA EWR and recall improvement bills. Markey also pressed Administrator Mark Rosekind to support a rulemaking to ensure EWR transparency.
What does the OIG report really tell us? Things may have gotten bad enough to constitute a tipping point.
Industry has had its way with NHTSA for a long time, telling it in various ways to stick it: outright lying about defects, failing to meet reporting requirements, taking a tone in their Part 573 Notices of Defect and Non-Compliance. You name it – any way they could disrespect the agency, they did it. And the major players found willing partners in the last NHTSA Administrator David Strickland, who suffered whiplash during his hasty and circular exit from government to industry, and DOT Secretary and technical illiterate Ray “Nobody-Cares-More-About-Safety-Than-Ray-LaHood” LaHood, who solved defect disasters with political deals that had zero impact on safety.
But back-to-back GM and Takata crises, so soon after Admiral LaHood (un)successfully steered the agency past the shoals of Toyota Unintended Acceleration, have necessitated a severe course correction. Administrator Rosekind appears to be well-suited to clean up their messes. He assured the senators that NHTSA had identified more than twice the number of process improvements as the OIG – a score of NHTSA 44 to the inspector’s 17. The agency accepted Inspector General Calvin L. Scovill III’s suggestions and had already knocked some of them off the list.
Rosekind said today that NHTSA needed to question its assumptions about the information automakers supplied it and “We need to question our assumptions.”
We dare not hope that such a radical notion, for example, might result in the chucking of the agency Bible of assumptions on unintended acceleration, the so-called Silver Book. But the OIG’s break-down of the breakdown in ODI’s operations is a good place to start.