Johns Hopkins study recommends narrow travel lanes

Narrow travel lanes, which benefit walkable cities because they provide more room for pedestrians, bicyclists, and landscaping, also do not contribute to automobile crashes, according to a new nationwide study from the Johns Hopkins Bloomberg School of Public Health.

According to the research team, this study, A National Investigation of the Impacts of Lane Width on Traffic Safety, sampled more than a thousand street sections from seven cities and conducted one of the most comprehensive data collections on geometric and street design characteristics of any comparable research to date. 

The study looked at diverse cities, including New York City, Salt Lake City, Denver, Miami, Washington DC, Philadelphia, and Dallas, with various urban forms and travel conditions. Highways and rural roads, presumed to benefit more from wider lanes, were excluded from the analysis.

“We found that the number of crashes does not significantly change in streets with a lane width of 9 feet compared to streets with lane widths of 10 feet or 11 feet, after controlling for cross-sectional and street design confounding factors such as posted speed limit, traffic volume, on-street parking, median type, number of lanes, bus stops, and similar sense of visual motions, most likely because the difference in lane width is not noticeable to drivers,” according to the researchers. “Our models confirm that in some cases (in the speed class of 30–35 mph), narrowing travel lanes is associated with significantly lower numbers of non-intersection traffic crashes and could actually contribute to improvement in safety.”

Narrow travel lanes may be safer because drivers operate more cautiously—that’s the theory expressed by these and prior researchers who have come to similar conclusions. 

Urbanists have been struggling to get narrower streets approved for more than 30 years, typically running up against a traffic engineering bureaucracy that demands wide lanes and “clear zones” that make walkable, mixed-use development difficult. The paper cited a 1997 study by new urbanist traffic engineer Peter Swift that looked at 20,000 accident reports in Longmont, Colorado, and concluded that 24-foot-wide streets are the safest. It was 11 years before a traffic engineering journal would publish the study, and the profession has generally ignored it. New Urban News, the precursor to Public Square, reported the results in 1997.

The Johns Hopkins team notes that reducing lane width in urban arterials provides more space for bicycle lanes, on-street parking, wider sidewalks, and landscaped buffers while reducing pedestrian crossing distances. Other benefits include minimizing construction and maintenance costs, and reducing land consumption, asphalt pavement, impervious surfaces, and the consequent effects of urban heat islands in cities. Narrow lanes promote walkability and the use of streets by all travel modes. 

“Narrowing travel lanes … is the easiest and most cost-effective way to accommodate better sidewalk and bike lane facilities within the existing roadway infrastructure,” the authors note.

The team also found that narrowing travel lanes “could have huge impacts on property values, business operation alongside the streets, and even could be the difference between the feasibility and successful delivery of a design project.”

The team notes that the most immediate candidates for lane width reduction projects are streets with 11-13 foot lanes in areas with speed limits of 20-25 and 30-35 mph. The study recommends that cities and states set 10-foot lanes as the default width in urban areas and let engineers justify wider lanes. Lane widths can be right-sized during routine repaving.

However, reducing lane widths is likely to have limited benefit if other street design changes are ignored, the researchers warned. Thus, the study promotes context-sensitive design in general. The researchers conducted extensive interviews with Departments of Transportation, and they were most impressed with the context classification system adopted by the State of Florida.  

This Florida system allows thoroughfares to be designed according to the rural-to-urban Transect, which justifies different designs for downtowns and walkable neighborhoods as opposed to rural highways. “Perhaps the most important takeaway from our interview with FDOT was their innovative context classification system that helps traffic engineers to differentiate between an arterial (or other road classes) in a low-speed (such as downtown) versus high-speed context.”

The team also noted progress in other states, like Vermont, which was the first state to adopt its own design standards rather than the widely used AASHTO Green Book, the study reports.

The study looks at the number, not the severity, of crashes and thus may underestimate the case for narrower travel lanes. Author and planner Jeff Speck makes a strong case for why the severity of crashes would also decline with narrow travel lanes. This argument is bolstered by case studies published by the Institute of Transportation Engineers in its 2017 book Implementing Context Sensitive Design on Multimodal Corridors, which found severe accidents often declined significantly when lane widths were reduced. Hamburg, New York, for example, found a 90 percent reduction in serious injuries on US Route 62, which travels through the village downtown. Over time, such a change would reduce health costs by millions of dollars and raise quality of life substantially by avoiding severe injuries.

The authors make some exceptions to their recommendations. High-speed highways, which are not covered in the study, benefit from 12-foot lanes. Regions with harsh winters also may need to avoid the narrowest streets due to snow removal considerations. “Our study does not recommend lane widths of less than 10 feet in the speed class of 20–25 mph and lane widths of less than 11 feet for the speed class of 30–35 mph in areas with harsh and heavily snowing winters,” the authors say.

Even with these exceptions, the case for narrower lanes in all urban areas keeps getting stronger, and the Johns Hopkins study, on top of previous research, will be hard for the traffic engineering profession to ignore. What is needed is a system that makes it easier for street designers to differentiate between walkable neighborhoods and rural highways. The Johns Hopkins researchers found the model for a successful context classification system in Florida.