THE NUMBERS: Annual deaths* from –
* 2010 estimates for world, 2012 for the United States
WHAT THEY MEAN:
Google tests a driverless car with no steering wheel; Honda introduces an auto-pilot-backed Accord to Australia; “Cruise” offers a $10,000 upgrade-to-autopilot feature for Bay Area Audis; Ford has its own plans. Auto-industry writer Dale Buss scoffs, writing in Forbes magazine that with auto fatalities trending down, at least on safety grounds there’s no need for anything radically new:
“It’s bogus to assert that driverless cars are required to address the issue of traffic accidents and fatalities. The number of U.S. auto-crash fatalities was declining steadily for years until very recently, when distracted driving, thanks largely to texting, reared its ugly head. But surely there are less drastic solutions to that development than removing every American driver from behind the wheel.”
True enough, auto fatalities have been heading down in the United States. Better manufacturing, safety regulations, and changing social attitudes on alcohol and seat-belts have cut the National Highway Traffic Safety Administration’s traffic death count rapidly and deeply, from the mid-1970s peak of 55,000 to 45,000 deaths in 2006, 32,885 in 2011 – the lowest number of traffic deaths since 1949 – and 33,561 in 2012.
So, significantly lower numbers. But “lower” shouldn’t be confused with “low.” An annual U.S. rate of 33,000 deaths (which includes 2,700 teenagers) is six times the annual number of deaths to HIV/AIDS, double the annual deaths to murder, and comparable to the casualty rates for the Korean War or World War I. Put another way, using state traffic figures, nine Texans die daily in car crashes, three each week Maine, one daily in Kansas, two per day in Illinois, and so on.
And worldwide, the U.S. is actually relatively safe. The World Health Organization’s Global Status Report on Road Safety 2013 estimates 1.24 million traffic deaths a year – nearly triple the UN’s estimate of annual murders, and twenty times the estimated annual total of deaths in wars. The toll is highest in middle-income countries, as a riskily picturesque mix of vehicles – cars, scooters, small motorcycles, trishaws, sometimes farm animals and carts – combines with the absence of helmet laws, and weaker police and emergency medical services. The worldwide traffic fatality rate is about 17 per 100,000 people – 50 percent above the 11.4 percent in the U.S., and likely higher if a world figure for deaths per vehicle-mile were available.
Why do all these crashes happen? NHTSA believes “human error is the critical reason for 93 percent of crashes.” A Missouri review of the state’s 716 fatal crashes in 2011 found equipment or road failures responsible for 11, while the other 705 involved human error – typically some combination of speeding, drinking, bad lane changes, and so on.
Now, back to the driverless car. Options range from Google’s experiment with the fully ‘autonomous’ vehicle to more evolutionary changes, in which an ‘autopilot’ guides most driving decisions, and the ‘Internet of things’ enables cars to sense approaching objects, scooters, people and stop by themselves. In principle, removing the driver can eliminate accidents resulting from drunkenness, drug use, texting, falling asleep, jumping red lights, looking left while a headphone-wearing pedestrian walks into your way from the right, and other human errors
No doubt, the concept raises engineering challenges, software questions, and marketing questions. Certainly after 120 years of eyes on the road and hands on the wheel, it is ‘radical.’ But assuming the rate of human error in fatal crashes is essentially the same as that in crashes overall, in principle driverless cars could cut deaths by 90 percent in the U.S.. Reductions worldwide would likely be smaller in percentage terms, since so many developing-world traffic deaths are in scooter, motorbike, or trishaw accidents. But even a reduction by half would save many more lives than the abolition of murder and war. In which case, a radical idea – even a “drastic” idea – might be a very good one.
Things to come –
Dale Buss in Forbes says don’t worry about driverless cars: http://www.forbes.com/sites/dalebuss/2014/06/30/google-jangles-auto-execs-but-will-driverless-cars-really-ever-happen/
Honda’s auto-piloted Accord makes its Australian debut: http://www.motoring.com.au/news/2013/medium-passenger/honda/accord/self-driving-honda-here-next-week-36348
Via Slate, Cruise’s Bay Area retrofitted autopilot for Audis: http://www.slate.com/blogs/future_tense/2014/06/24/cruise_s_10_000_self_driving_kit_will_come_to_market_before_anything_driverless.html
Google’s self-driver car video: http://www.google.com/about/careers/lifeatgoogle/self-driving-car-test-steve-mahan.html
And Ford’s Blueprint for Mobility: http://corporate.ford.com/microsites/sustainability-report-2011-12/financial-mobility-blueprint
The World Health Organization’s Global Status Report on Road Safety 2013 tabulates 1.24 million deaths worldwide: http://www.who.int/violence_injury_prevention/road_safety_status/2013/en/
WHO finds American roads a bit more dangerous than those in other rich countries, with a rate of 11.4 traffic deaths per 100,000 people – a level comparable to upper-middle income countries like Poland and Chile. (This could overstate U.S. risk, as the Report covers only deaths per capita, rather than per vehicle-mile.) The world’s lowest traffic-fatality rate are in a few small island countries – the Maldives and Micronesia in particular – where few people own cars. Setting them aside as special cases, the lowest rate is Iceland’s 2.8 per 100,000; others include the U.K.’s 3.7 per 100,000, Germany’s 4.7, and Japan’s 5.2.
The highest rate in the world is the Dominican Republic’s 42 deaths per 100,000 people. This attributable in particular to very high and unsafe use of motorcycles: 57% of DR traffic deaths are drivers of, or passengers on, motorcycles or scooters; 25% are pedestrians, and 14% are automobile drivers. Other high rates include 38 deaths per 100,000 in Thailand (74 percent are tuk-tuk or motorcycle drivers or passengers); 37 per 100,000 in Venezuela and Iran, and 34 per 100,000 in Nigeria.
The National Highway Traffic Safety Administration on American traffic fatalities, up to 2012: http://www.nhtsa.gov/NCSA
And the CDC on the awful toll of teenage driving accidents:
The U.N.’s Office of Drugs and Crime estimates murders for 2012 at 437,000: http://www.unodc.org/gsh/
And Oslo’s Peace Research Institute finds battle deaths now averaging 55,000 per year: ”http://www.prio.org/Data/Armed-Conflict/Battle-Deaths/The-Battle-Deaths-Dataset-version-30/
And two stories from abroad –
AP looks at motorcycles and tragedy in the Dominican republic: http://bigstory.ap.org/article/dominican-traffic-death-rate-among-worlds-highest
And the Hanoi-based Asia Injury Foundation looks at traffic deaths – about 12,000 Vietnamese, mostly frequently young scooter-owners driving without helmets, die each year in traffic accidents – in the developing world, helmet law policies, and other options: http://asiainjury.org/the-challenge/