Internet Explorer 11 is not supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

Speeding Plays an Even Bigger Role in Traffic Deaths Than We Thought, Say Feds

The National Transportation Safety Board wants governments to crack down on speeding, which claims as many traffic deaths as drunk driving. But the hard question is: How?

(Flickr/Brandon Harer)
A new study out of Washington is rarely a cause for celebration, but many traffic safety groups are excited about a forthcoming report that highlights the big role speeding plays in traffic deaths.

The study comes from the National Transportation Safety Board, an agency best known for its investigations of deadly plane crashes and train derailments. The NTSB has also been a force behind safety innovations, like air bags in cars and graduated driver’s licenses for teen drivers.

Researchers have actually underestimated how often speed is a factor in fatal crashes, according to a summary of the report, which will be released in full in coming weeks. That’s significant, considering that speed is already one of the most widely reported causes of deadly crashes. In 2015, for example, it was identified as a factor in roughly as many traffic deaths (9,557) as alcohol (9,306) or people not wearing seat belts (9,874).

But the NTSB went further, by urging traffic engineers to rethink how they set speed limits and for states and localities to use speed cameras more often. NTSB wants law enforcement agencies to mount a national anti-speeding campaign, akin to “Click It or Ticket” for seatbelt use. The agency also wants carmakers to install features to alert drivers when they’re going over the speed limit and maybe even slow them down automatically.

It’s an ambitious plan, requiring major changes by private industry and by all levels of government. But safety advocates welcomed the broad-based approach. “The biggest thing for us was the way that the report highlights speed as big of a problem as it is. It’s often something that’s overlooked,” says Russ Martin, the director of government relations for the Governors Highway Safety Association. “It’s a traffic safety problem on par with drunk driving, and we hope that can dedicate resources to preventing speeding the same way we do that for drunk driving.”

Emiko Atherton, the director of the National Complete Streets Coalition, is one of the authors of that group’s annual “Dangerous by Design” reports, which show how street design encourages drivers to go too fast and cause fatal or harmful crashes. Advocates such as Atherton have been talking about the role of speed in crashes for a long time, but they don’t have the national reach that NTSB does, she says. “For this nationally appointed board to come out and say, ‘Speeding is one of the primary factors in traffic fatalities,’ is huge.”

Indeed, advocacy groups have long been concerned about speeding, and those concerns have only grown in recent years.

The United Nations declared a “Decade of Action for Road Safety” in 2011 and later specified that it wants to reduce worldwide road deaths by 50 percent by 2020. One of its main emphases is reducing speeds. “Speed is at the core of the road traffic injury problem,” the World Health Organization wrote.

Closer to home, the uptick in roadway deaths in the last two years has spurred interest in approaches like Vision Zero, which hopes to end traffic fatalities. The approach relies heavily on redesigning streets to slow vehicle speeds where other users, like cyclists, pedestrians or transit riders, are present.

But the shift also comes as traffic engineers and others challenge, or at least re-examine, assumptions about travel speed and safety that have shaped their policies for decades.


The Dangers of Speeding

In 1964, researcher David Solomon published a study for the federal government that is relied on to this day. Solomon concluded that crash rates were highest at very low speeds, lowest at average speeds and higher again at above-average speeds. “Thus,” he wrote, “the greater the variation in speed of any vehicle from the average speed of all traffic, the greater its chance of being involved in an accident.”

So, to reduce crashes, engineers designed roads that move traffic efficiently, without a lot of stop-and-go that would increase the speed differential between vehicles. The same philosophy affected how state and local governments set speed limits. The federal government recommends that officials rely on the “85th percentile” rule to set speed limits, meaning the limits should be high enough that 85 percent of drivers travel under the speed even when no speed limit signs are posted.

Solomon’s research, though, had serious limitations. Most glaringly, it was all conducted on rural highways (but not freeways). And Solomon’s report did not examine alcohol use or seatbelt use, which have drawn significant attention in the half-century since. His report looked at road design, but only when it came to road widths and the number of intersections and driveways per mile (which, he said, showed the “benefits of controlling access to the highway”). Finally, Solomon’s report also focused exclusively on people traveling in vehicles, not pedestrians or cyclists that also use the roadway.

So this summer, NTSB’s researchers echoed what many safety experts have been saying: It’s hard to determine whether speed causes crashes, as Solomon suggested, because of other factors that might be at play too, such as the road type, driver’s age, alcohol and the design of the road. But what is clear, as even Solomon pointed out, is that speed has a big impact when crashes do occur. “The relationship between speed and injury severity is consistent and direct,” the NTSB researchers wrote.

In fact, a 5 percent cut in average traffic speed can result in a reduction of 30 percent in the number of fatal crashes, according to the World Health Organization.

The NTSB pointed out that high speeds were “especially critical for pedestrians,” because they have no protection in vehicle crashes. That echoes findings from the AAA, the automobile association, in a 2011 report, which showed how pedestrians’ risk of death from a vehicle crash rose dramatically as the speed of the vehicle increased by seemingly small increments.

The risk of a pedestrian dying is 10 percent when the vehicle striking them is going 23 mph; if the vehicle is going 32 mph, the walker has a 25 percent chance of dying. Half of pedestrians die when the car or truck is going 42 mph, three-quarters die when vehicles are traveling 50 mph and 90 percent die when vehicles reach 58 mph.


A Push for Speed Cameras

One of NTSB’s most controversial recommendations is to increase the use of speed cameras, which are often unpopular with the public and met with skepticism by state lawmakers. Seven states ban the cameras altogether, 15 restrict their use and 28 have no law explicitly authorizing the automated enforcement. Several states also limit the amount of revenue cities can collect from speeding fines.

All told, 142 jurisdictions in 14 states and the District of Columbia use speed cameras, which is only a third of the number that use red-light cameras.

The federal safety researchers argued that states ought to explicitly authorize speed cameras. They say limiting where cameras can be used to school zones or work zones, for example, also limits the effectiveness of the devices in making roads safer.

Studies show that speed cameras do reduce both speeds and accidents, the NTSB researchers noted. They pointed to a 2010 analysis of 28 speed camera deployments. In every case, using speed cameras reduced crashes. The size of the reduction varied from 8 percent to 49 percent, while severe crashes that caused serious injuries or deaths dropped 11 percent to 44 percent.

Martin, from the governor’s highway safety group, says officials are getting better at developing speed camera policies that avoid the types of controversy that have surrounded them in the past. One key to that is making sure that the traffic cameras are being used to promote safety more than to raise revenue, he says. Governments need to publicize that they are using the cameras and emphasize that they want to deter people from speeding.

“It’s just a question of getting all of the details right and making the case to the public that these programs are here for a reason, and we’re doing the right thing,” Martin says.


Setting Speed Limits

But enforcement only goes so far, especially if the speed limits on roads are too high. Since 2012, though, states largely have been moving to raising speed limits, rather than decreasing them. In 2012, only Utah and Texas had top speed limits of 80 mph or more, but by 2016, five more states joined them. By comparison, in 2012, 16 states had top speed limits of 55 to 65 mph; by last year, that number fell to 10.


Maximum Speed Limits: 2017
65 mph or less  
70 mph  
75 mph  
80-85 mph

SOURCE: IIHS; current as of August 2017
Maximum Speed Limits: 2012
65 mph or less  
70 mph  
75 mph  
80-85 mph


SOURCE: GHSA "Speeding and Aggressive Driving"; published March 1, 2012
NOTE: States set different limits for urban roadways, rural roadways and limited access roads. Some also allow speed limits to be set higher on limited segments if traffic studies support it. Maximum speed limits shown are the single highest limits among any type of roadway within a state.
At the other end of the spectrum, many cities, like New York City, don’t control the speed limits on their own city roads; the state does. So when New York wanted to lower the citywide speed limit from 30 mph to 25 mph, it had to lobby Albany to make the change.

But NTSB took aim at another problem with how speed limits are set: the commonly-used 85th percentile rule inspired by Solomon’s 1964 study. (Some jurisdictions set speed limits by law. A city, for example, can set a citywide speed limit for residential streets, a different one for main thoroughfares and a third limit for highways. The 85th percentile rule, though, only applies in areas where jurisdictions rely on engineering studies to set their speed limits.)

“Raising speed limits to match the 85th percentile speed can result in unintended consequences,” the NTSB researchers wrote. One of those problems is that, by setting the speed limit so high, it can actually encourage drivers (most of whom would otherwise go slower than the new speed limit) to go even faster. That effect can even last once drivers leave the area with the new speed limit.

Plus, the researchers said “there is not strong evidence” that setting the speed limit at the 85th percentile actually results in the lowest crash rates.

In effect, the 85th percentile method assumes that most motorists are reasonable and drive prudently to avoid crashes and dangerous situations. But research (and most drivers’ experience) shows that’s not always the case.

Despite the drawbacks, the approach does have many benefits. A 2012 Federal Highway Administration report explained that the method is “also attractive because it reflects the collective judgment of the vast majority of drivers.” In general, laws should not make people acting reasonably into law-breakers. “Setting a speed limit even 5 mph below the 85th percentile speed can make almost half the drivers illegal; setting a speed limit 5 mph above the 85th percentile speed will likely make few additional drivers legal.”

So the NTSB suggested that the federal government change its rules for how states and localities set speed limits. Jurisdictions that set their speed limits with engineering studies should take into account, not only the average speed of motorists, but also the conditions of the road, development along the road, parking, the presence of pedestrians and the crash history of the area.


What About Road Design?

Even stricter speed limits might not be enough. Jeff Lindley, the associate executive director for Institute for Transportation Engineers, says motorists can feel that speed limits are arbitrary if the design of the road doesn’t match the speed limit. A straight road with wide lanes and a median signals to drivers that they can go fast, even if there’s a lower speed limit to protect pedestrians or other vulnerable users of the road. If you give drivers a ticket on that type of road, they’ll likely say they are going a reasonable speed and just keeping up with traffic, he says.

So, increasingly, traffic engineers are trying to design roads that reflect the needs of all users, not just motorists. “The design of a facility can help send the message of what the proper speed is and encourage people to drive at that speed rather than a faster speed,” Lindley says.

The NTSB report did not explore the issue of road design, and that’s a missed opportunity, says Atherton, the director of the National Complete Streets Coalition. “You have to pair speed limits with physical traffic-calming measures for them to be effective,” she says. “Just lowering the speed limits is insufficient.”

One of the NTSB commissioners asked the agency’s researchers during their presentation why road design wasn’t emphasized in the report. One of the authors said that other publications, like street designs by the National Association of City Transportation Officials (NACTO) and the Federal Highway Association already explained in great detail how to improve road design to improve safety.

And Lindley, from the Institute for Transportation Engineers, says there’s a growing awareness in the engineering industry of the need to design roads for all users, which frequently requires including traffic calming features. That’s especially true for new projects, even though many dangerous existing roads, like suburban arterials, will be harder to retrofit.

Last year, in fact, the ITE began embraced the goal of achieving zero traffic deaths and began promoting Vision Zero, which encourages changes to roadways to slow down traffic. The fact that traffic fatalities started climbing, after years of decreases, drove the point home.

“[Fatalities] are at or approaching 40,000. That’s a huge number. It’s an unacceptable number,” he says. ”There is a recognition that what we’ve been doing to focus on safety [is] not having the desired effect. That opens the door to doing something different. Vison Zero is a very different approach to safety.”

From Our Partners