Are Traffic Deaths Really Up?

A headline in the Feb. 16, 2017 New York Times read "U.S. Traffic Deaths Rise for a Second Straight Year".

In quoting the National Safety Council, the source of this grim claim, writer Neal E. Boudette wrote: "For the first time since 2007, more than 40,000 were killed in crashes last year, a safety group estimated, pointing to lax law enforcement as a factor."

The story goes on to say that "...government officials and safety advocates contend... the increase in deaths has been caused by more lenient enforcement of seatbelt, drunken driving and speeding regulations by authorities and a reluctance by lawmakers to pass more restrictive measures."

The New York Times wasn't alone in sounding the alarm. Dozens of its media contemporaries weighed in with similar stories. And without exception, they all got it wrong.

For starters, that 40,000-fatality claim by the NSC was merely a guess. The fatality numbers for 2016 won't be available until the National Highway Transportation Safety Agency issues its report later in 2017.

There are other problems behind that headline-grabbing number. For example, the National Safety Council counts traffic deaths that occur within a year of the accident and also includes accidents that occur off public roads.

In contrast, NHTSA counts only  traffic deaths on public roads and which occur within 30 days of the accident.

With these expanded criteria, the NSC's fatality numbers are larger and less accurate than those from NHTSA. Buried on its website the NSC admits as much: "National Safety Council figures are not comparable to National Highway Traffic Safety Administration figures."

The National Safety Council routinely issues the same sky-is-falling press release every year.  A February 17, 2016 NSC press release screamed "Motor Vehicle Deaths Increase by Largest Percent in 50 Years".

"Preliminary estimates from the National Safety Council indicate motor vehicle deaths were 8% higher in 2015 than they were in 2014 - the largest year-over-year percent increase in 50 years," the release read.

"The Council estimates 38,300 people were killed on U.S. roads, and 4.4 million were seriously injured, meaning 2015 likely was the deadliest driving year since 2008".

Wrong again. NHTSA reported 35,092 fatalities for 2015, 9.1 percent lower than the NSC's shrill prediction.

To compare year-to-year changes the death rate is the critical number. It's expressed in fatalities per 100 million Vehicle Miles Traveled (VMT).

In 2015 there were 1.12 traffic fatalities per 100 million Vehicle Miles Traveled. Statistically, this means you'd need to drive 89.3 million miles before your number comes up.

What about the National Safety Council's claim that 2015 was the "deadliest driving year since 2008"?

Hardly. 2015 saw the second-lowest fatality rate since NHTSA began keeping records. It slightly trailed the 1.08 rate of 2014, the all-time low.

By rights the Times piece should have read "U.S. Traffic Safety Nears All-Time High". But statements like that don't grab headlines or sell newspapers.