Newly released estimates from the National Highway Safety Administration (NHTSA) indicate that traffic fatalities jumped by nearly 12 percent during the first nine months of last year compared to the same period in 2020. That's "the highest percentage increase during the first 9 months since NHTSA started collecting fatality data in 1975," notes a press release from Mothers Against Drunk Driving (MADD). The headline: "New Report Reveals Staggering Increase in Traffic Fatalities on U.S. Roadways in First 9 Months of 2021."
Notably missing from MADD's press release: any mention of the COVID-19 pandemic, which in 2020 led to a sharp reduction in driving, thanks to a combination of voluntary precautions and legal restrictions. Driving has rebounded since then, which makes a crucial difference in assessing the change highlighted by MADD. Far from showing a "staggering increase," the estimated fatality rate per 100 million vehicle miles traveled (VMT) was essentially the same during the first nine months of 2021 as it was in the first nine months of 2020: 1.36 vs. 1.35.
You might dismiss MADD's framing of NHTSA's numbers as the predictable hyperbole of an advocacy group looking for ammunition in policy debates. But press coverage of the increase in traffic deaths also conspicuously overlooked the most obvious explanation.
There is legitimate cause for concern about longer-term trends in traffic fatalities. The annual fatality rate per 100 million VMT rose by 23 percent in 2020, from 1.11 to 1.37, after declining for three consecutive years. But focusing on the change between 2020 and 2021 while emphasizing total numbers rather than rates is more than a little misleading.
Between February 2020 and February 2021, VMT during the previous 12 months fell by more than 13 percent. By contrast, NHTSA reports that "vehicle miles traveled in the first nine months of 2021 increased by about 244 billion miles, an 11.7% increase from the same time in 2020." That is about the same as the increase in crash deaths, which is why the fatality rate barely budged.
MADD is using the new NHTSA numbers to reinforce its argument for "swift implementation of the Congressionally mandated rulemaking to require drunk driving prevention technology in all new cars." Whatever the merits of that policy, the fact that the traffic fatality rate remained about the same last year is obviously not helpful from MADD's perspective, which presumably is why it chose to ignore that important point. But the Associated Press story about NHTSA's report also plays up the increase in total fatalities.
"US road deaths rise at record pace as risky driving persists," the headline says. "The estimated figure of people dying in motor vehicle crashes from January to September 2021 was 12% higher than the same period in 2020," A.P. reporter Hope Yen writes. "That represents the highest percentage increase over a nine-month period since the Transportation Department began recording fatal crash data in 1975."
Unlike MADD, Yen does mention the pandemic, but without noting its impact on miles traveled. Instead she reports that "NHTSA has blamed reckless driving behavior for increases during the pandemic, citing behavioral research showing that speeding and traveling without a seat belt have been higher." An October 28 A.P. story about NHTSA's estimates for the first six months of 2021, by contrast, noted that "the latest spike in fatalities came as people drove more as pandemic shutdowns eased." It cited data indicating that "vehicle miles traveled in the first six months of the year rose by 173.1 billion miles, around a 13% increase from last year."
The idea that drivers became more reckless in 2020 is supported by NHTSA's research, and it is consistent with the increase in the annual fatality rate per 100 million VMT between 2019 and 2020. The 2020 rate was the highest reported since 2007, although still much lower than the rates seen in the preceding decades. While it looks like that relatively high rate persisted during the first nine months of 2021, NHTSA notes that "the fatality rates in the second and third quarters of 2021 declined compared to 2020."
The real story here is not a "staggering increase" in total traffic fatalities between 2020 and 2021. It is the even larger increase in the fatality rate between 2019 and 2020, which was a stark contrast with the remarkable progress seen since the 1970s, '80s, and '90s. The 2020 rate was still less than one-third the 1970 rate, less than half the 1980 rate, 37 percent lower than the 1990 rate, and 10 percent lower than the 2000 rate. But it is reasonable to worry that the 2020 increase signals a problem that will persist after the pandemic.
According to NHTSA's research, "average speeds increased during the last three quarters of 2020, and extreme speeds, those 20 miles per hour (or more) higher than the posted speed limit, became more common." At the same time, "analyses
of data from fatal crashes" showed "an estimated 11% increase in speeding-related fatalities." NHTSA says the explanation for those changes is not entirely clear: "While previous research had posited that decreases in VMT in 2020 had allowed increases in speeds compared to 2019, the return of VMT in 2021 paired with increased speeds on different roadway types in 2021 suggests this supposition may not completely explain differences in behavior."
Other data "suggested fewer people in crashes used their seat belts." Ejections per 100 crashes spiked in the first quarter of 2020, then fell but remained at a higher level than usual for the rest of the year.
An October 2020 NHTSA study found that 27 percent of seriously or fatally injured roadway users tested positive for alcohol in the spring and summer of 2020, up from about 21 percent before the pandemic. The share testing positive for opioids rose from less than 8 percent to nearly 13 percent. According to survey results reported last September, 7.6 percent of U.S. drivers said they were more likely to drink and drive during the pandemic. NHTSA adds that "increases in sales of alcohol and marijuana, while indirect measures of risk to road traffic safety, are indicators of social changes that could have traffic safety implications."
Assuming that the social and economic disruption caused by COVID-19 explains the increases in risky behavior described by NHTSA, the traffic fatality rate may continue the downward trend observed in the second and third quarters of 2021. In a year or two we should have a better idea of what is happening.
"Transportation Secretary Pete Buttigieg has pledged help and released a new national strategy last week aimed at reversing the trend, which he calls a crisis," A.P. reports. His solution includes "billions in grants under President Joe Biden's new infrastructure law to spur states and localities to lower speed limits and embrace safer road design such as dedicated bike and bus lanes, better lighting and crosswalks." Buttigieg "also urges the use of speed cameras, which the department says could provide more equitable enforcement than police traffic stops."
Those policies, like the ones favored by MADD, should be judged on their merits. But it would help to understand the nature of the "crisis" that Buttigieg perceives. At this point, it's not clear whether the 2020 increase in the traffic fatality rate represents a lasting reversal or a pandemic-related anomaly.