Cars

A Safer Road Map Toward Self-Driving Cars

Auto deaths are rising for the first time in decades, and lax regulations on autonomy won't help.

The future isn't quite here.

Photographer: Robin Van Lonkhuijsen/AFP/Getty Images

Ever since then-candidate Donald Trump began naming and shaming automakers for outsourcing American manufacturing jobs, the automobile industry has figured heavily in the new president’s political program. Having changed the industry’s tone -- if not its actions -- around manufacturing jobs in the course of a few tweets, Trump now appears to be rewarding the industry by relaxing tough 2025 fleet fuel economy standards set by the Obama administration.

There may be something to this: By alternately confronting and conciliating the auto industry, Trump is plotting a maverick course that one hopes could break up the longstanding cozy relationship between carmakers and the government agencies that oversee them.

Now, with a host of futuristic automotive technologies hitting American roads -- from lithium-ion batteries and hydrogen fuel cells to autonomous drive systems and shared-mobility apps -- the people Trump picks as top auto safety officials will play a momentous role in guiding America’s transition toward bewildering new transportation options.

But before even beginning to future-proof America’s auto safety regulations, the Trump administration will have to tackle a more immediate and prosaic challenge. For the first time since the National Highway Traffic Safety Administration was created in 1970, on-road deaths have increased for two years running. With 40,000 people dying on American roads in 2016 alone -- and also a recent rash of shocking auto safety scandals involving Takata Corp. airbags and General Motors Co.’s ignition switches -- auto safety is a populist issue just waiting for the kind of oxygen Trump can provide.

About half of fatal crashes each year involve an occupant not wearing a seatbelt, while nearly a third involves a drunk or otherwise impaired driver. A model solution here is the Obama administration’s mandate requiring a back-up camera in every new vehicle by 2018. Admittedly, that rule, which only addresses some 200 driveway tragedies per year, was expensive for carmakers, which passed most of the costs along to buyers.

But mandating seatbelt and breathalyzer ignition interlocks -- which keep the car from starting until the driver and passengers comply -- would be far less expensive. And thanks to improvements in computer and sensor technology, these devices are no longer the nuisance they were when a seatbelt interlock mandate was briefly in place in the 1970s. In fact, automakers including BMW are actively lobbying the NHTSA to bring belt interlocks back.

By directly tackling the most common causes of road death with simple but effective regulations like seatbelt interlocks, NHTSA could also free up much of the hundreds of millions of dollars it spends each year on highway traffic safety grants, most of which are ineffectually aimed at “nudging” the public and state governments toward wider use of seatbelts and reductions in intoxicated driving. These grants make up more than half of the agency’s budget, so judicious cuts enabled by new interlock mandates could free up the resources to improve its core data-collection, investigation and enforcement functions.

Of all the new technologies threatening to upend the century-old institution of the automobile, autonomous drive is by far the most significant for auto safety regulators. Under the Obama administration, the NHTSA embraced the long-term potential for autonomous cars to vastly improve safety, setting a goal that the technology could reduce road deaths to zero within 30 years. Unfortunately, for all the exciting potential that self-driving cars show, NHTSA’s techno-optimism has made it too lax in addressing the disturbing safety risks that nascent autonomy currently presents.

Though experts are divided on how far off true autonomous car technology is, they agree that current systems such Tesla Inc.’s Autopilot and Mercedes-Benz AG’s DrivePilot fall well short of true autonomy. That has not prevented the companies from aggressively pushing untested “public beta” capabilities onto public roads, while implying or even stating that these systems are already “self-driving.” At least one American has been killed in a crash while driving in autopilot mode. Yet the NHTSA seems not to appreciate the extent to which these systems can lull drivers into a false sense of security, and has exonerated Tesla in the fatal crash, despite troubling questions about its key findings.

The laxity with which federal safety regulators have approached autonomous cars more broadly has now created a regulatory vacuum. To date, the Federal Department of Transportation has come up with only policy “guidelines” for autonomous vehicles, while state and local regulators are wan to set clear rules, seemingly in a competition to create the friendliest possible environment in hopes of attracting investment. Already there is no shortage of evidence that some companies expect to be able to defy even these weak local rules -- as Uber Technologies Inc. and its Otto division have in California and Nevada

Developing comprehensive regulations that strike a balance between allowing companies to innovate and protecting public motorways will be a lengthy and difficult process that will extend well past the Trump presidency. DOT’s policy guidelines are not a bad roadmap for the journey. But because they have no teeth, we are headed toward a culture of impunity among the most aggressive firms vying for self-driving domination, which tougher new state-level regulations like California’s will struggle to check.

The long-term safety advantages of autonomous cars do not excuse putting Americans at risk in the short term. A new ad-hoc federal policy centered on ramping up monitoring of autonomous car testing, investigation of each report of dangerous activity, and the threat of meaningful penalties for irresponsibility, would make it clear that the NHTSA will not tolerate companies who put the public at risk. And if well-designed, a smart federal monitoring regime could actually make it easier for companies to comply with the wide variety of inconsistent and overlapping local regulations.

Yes, we need to be wary of excessive rules that crush innovation. But it’s possible to come up with relatively simple measures that directly address the root causes of most road deaths, satisfying safety advocates and limited-government types alike. There will be willing corporate partners among the automakers who support the safety benefits of seatbelt interlocks and the technology companies who worry that an irresponsible autonomous-car gold rush could turn the public against their massive investments in technology.

The core of this new deal is the promise that huge amounts of ineffective and overlapping regulation -- governing both traditional auto safety and new autonomous safety challenges -- can be replaced by fewer, and more consistently enforced, regulations aimed directly at the heart of the problems. By forging an agenda that satisfies so many seemingly irreconcilable interests, Trump has the opportunity to save lives, improve government and build a new consensus around the inevitable future of transportation.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

    To contact the author of this story:
    Edward Niedermeyer at eniedermeye1@bloomberg.net

    To contact the editor responsible for this story:
    Tobin Harshaw at tharshaw@bloomberg.net

    Before it's here, it's on the Bloomberg Terminal.
    LEARN MORE
    Comments