The Magic Bullet of Road Design: Narrower Lane Widths

It isn’t often that a complex problem can be significantly solved by a single remedy.  But when it comes to finding ways to make car-dominated streets more pedestrian and cyclist-friendly, narrowing the lane widths is a game-changer.  Critics worry about safety and capacity, but new research refutes these fears. 

It is for good reason that traffic engineers tend to be conservative.  They understand that road design shapes driver behavior and bad design leads to confusion, accidents, and injuries – even deaths.  When an engineer signs off on a design, he (and occasionally she) is putting his professional future on the line, if not exposing himself to potential liability claims.  You avoid trouble by following proven established practice; being innovative or non-standard means increasing risk.

The traffic design “bible” is AASHTO’s “Green Book” (the most recent edition is officially titled Geometric Design of Highways and Streets, 2004, published by the American Association of State Highway and Transportation Officials).  The Green Book calls itself a “guide” – meaning that they are simply suggestions – but many states and municipalities treat AASHTO recommendations as required standards.  Even so, while AASHTO publications tend to suggest ranges for key road dimensions, with the exact size to be chosen dependant on the particular context of each project, professional practice is to use the maximum size whenever possible.  This is partly because of the tendency to “overbuild” by designing roads to handle faster speeds, heavier traffic, and bigger vehicles than officially anticipated – partly to provide an extra margin of safety, partly to prepare for possible growth, and partly because of the “cars are the future” attitude that has dominated American planning and culture for much of the past century.

For lane widths, AASHTO suggests that “local” roads be between 9 and 12 feet wide, that “collectors” be between 10 and 12 feet wide, and that “arterials” be between 10 and 12 feet wide (between 11 and 12 feet in rural areas).   The Federal Highway Administration’s publication, “Mitigation Strategies for Design Exceptions” (July 2007, chapter 3) states:  “Narrower lane widths may be chosen to manage or reduce speed and shorten crossing distances for pedestrians. Lane widths may be adjusted to incorporate other cross-sectional elements, such as medians for access control, bike lanes, on-street parking, transit stops, and landscaping. The adopted ranges for lane width in the urban, low-speed environment normally provide adequate flexibility to achieve a desirable urban cross section without a design exception.”

However, many state, municipal, and corporate-employed traffic engineers won’t sign off on any anything less than 11 or even 12 feet.  Given the narrow and unchanging distance between buildings on many urban streets, going for the maximum traffic lane widths leaves no room for widening sidewalks, installing bicycle lanes or cycle tracks, putting in median or intersection “bulb outs,” replacing storm sewers with more environmentally-friendly water runoff methods, adding street furniture or trees or other aesthetic improvements, and much else that would make streets more attractive and useful.

Fortunately, recent research has proven that narrower lane widths neither reduce safety nor lower through-put.  In fact, they may even be safer!  Unfortunately, while official policy is slowly beginning to recognize these facts, road design practice and construction still resists.  As a result, too many roads are still being built or rebuilt as if the only thing that mattered was car traffic.

Some Research Results Concerning Safety of Narrower Lanes

“Wider traffic lanes provide give vehicles more room to avoid collisions, but they increase traffic speeds, which increases crash risk and severity. Annual crash rates per vehicle-mile tend to be lowest for relatively narrow (about 10-foot) lane widths, and are highest on wider, lower volume, straight streets with higher traffic speeds.” (Swift, 1998; Zegeer, et al, 1994; CTRE, 2006).

“For multilane urban arterials and multilane rural arterials, the expected difference in substantive safety for variations in lane width is much less, on the order of a few percentage points, when comparing lane widths of 10 to 12 feet.”  (FHWA, Mitigation Strategies for Design Exceptions – July 2007, Chapter 3)


“All projects evaluated during the study that consisted exclusively of lane widths of 10 feet or more resulted in accident rates that were either reduced or unchanged.” (NCHRP 330, Effective Utilization of Street Width on Urban Arterials, Transportation Research Board, 1990.

“A safety evaluation of lane widths for arterial roadway segments found no indication, except in limited cases, that the use of narrower lanes increases crash frequencies. The lane width effects in the analyses conducted were generally either not statistically significant or indicated that narrower lanes were associated with lower rather than higher crash frequencies…. There is no consistent, statistically significant relationship between lane width and safety for midblock sections of urban and suburban arterials. There is no indication that the use of 3.0- or 3.3-m (10- or 11-ft lanes), rather than 3.6-m (12-ft) lanes, for arterial midblock segments leads to increases in accident frequency….There is no consistent, statistically significant relationship between lane width and safety for approaches to intersections on urban and suburban arterials. There is no indication that the use of 3.0- or 3.3-m (10- or 11-ft lanes), rather than 3.6- m (12-ft) lanes, for arterial intersection approaches leads to increases in accident frequency.” [emphasis added] (Ingred B. Potts, Harwood, D., Richard, K., Relationship of Lane Width to Safety for Urban and Suburban Arterials, Transportation Research Board, 2007 Annual Meeting.)

“As more arterial and collector lane widths are increased up to 12 ft or more, traffic fatalities and injuries increase….These results are quite stunning as it is general practice to ‘improve’ the safety of roads by increasing lane widths.”  (research presented by Dr. Robert B. Noland at the Transportation Research Board’s annual meeting in 2008)

Some Research Results Concerning Capacity of Narrower Lanes

The “official” Highway Capacity Manual” issued by the Transportation Research Board in 2000 claims that the capacity (saturation flow rate) of a lane at a signalized intersection is reduced by 3.33 percent for each foot of lane width less than 12 feet.  A 10 foot lane should only be able to carry 93 percent of the traffic of a 12 foot lane.

However, the city of Cambridge, Massachusetts put the Central Square section of Massachusetts Avenue, the city’s heaviest travelled street, on a “road diet” cutting it from 4-5 lanes down to 2-3 (width varies along the length), reprioritizing the space for pedestrians, bicyclists, and transit.  Despite this drastic reduction of official capacity, by carefully synchronizing traffic lights the city was able to maintain throughput at about 21,000 ADT.   In other words, allowing cars to go fast merely speeds up their arrival at the next red light while maintaining a steady, even if slow, forward motion ends up moving just as many cars while increasing safety for everyone.

Also, in 2007, the Florida Department of Transportation had a literature search done as part of its “Conserve By Bike Program Study which found: “The measured saturation flow rates are similar for lane widths between 10 feet and 12 feet. For lane widths below 10 feet, there is a measurable decrease in saturation flowrate. Thus, so long as all other geometric and traffic signalization conditions remain constant, there is no measurable decrease in urban street capacity when through lane widths are narrowed from 12 feet to 10 feet”. [emphasis added] (John Zegeer, P.E., [past Chair, TRB Jighway Capacity and Quality of Service Committee] in a memo to Sprinkle Consulting Engineers, March 22, 2007.)


Other Implementations of 10-Foot (or narrower) Lanes

A quick survey of members of the Association of Pedestrian and Bicycle Professionals (cited in “Ten-Foot Lanes for Boston” by Christopher Longenbaker in a paper for Professor Peter Furth of Northeastern University) reports the following examples of 10 foot lanes used on urban arterials:

  • • In Washington DC, many arterials have 10 foot lanes are also heavily used routes for bus and truck traffic. Examples include 16th Street, Connecticut Ave, Wisconsin Ave, Pennsylvania Ave, Florida Ave, U Street, 14th Street.
  • • In Chicago, North Sheridan Rd, with 42,000 ADT, and N. Ridge Ave, with 50000 ADT, have four 10 foot lanes that work well, even while carrying heavy bus and truck traffic.
  • • Florida DOT striped some lanes less than 10’ on US-41 in the 1980s. There was no difference in crash rates compared to 12 foot lanes.
  • • Missouri DOT restriped I-44 to 10 foot lanes, and there have been no reports of problems.
  • • Other cities that are using 10 foot lanes on arterials, including major arterials with trucks and busses, include Arlington, VA; Cincinnati; Colorado Springs; Charlotte, NC; Eugene, OR; Houston; Lawrence, NJ; several cities near Los Angeles; Portland, OR, Rochester, NY; San Jose; Scottsdale, AZ; Tucson; and Philadelphia. In several of these cases, lanes were narrowed to 10 feet in order to make space for bike lanes.

In the Boston area, LivableStreets Alliance compiled a list of some 10-foot wide arterials, all of which carried trucks and buses, including: Berkeley  and Claredon Streets in the Back Bay; Hampshire Street, Broadway in Kendal Square, and Mass Ave in Porter Square in Cambridge (the later two having 10.5-foot lanes); Beacon Street in Somerville, The BU, Mass Ave, and Craigie (Science Museum) bridges, Alewife Brook Parkway on the Cambridge/Somerville border, and Memorial Drive along the Charles River.

Research-based Specifications

Based on this research, advocates have begun pushing for road designers to use ASSHTO’s minimum suggested dimensions, and in some cases to experiment with pushing beyond.  Using the Complete Streets approach of starting from the outside and working inwards towards the center line, the following are the starting point for discussion:

  • Sidewalks – in residential areas, at least 5 feet wide to allow two people walking side-by-side or passing, but really as wide as possible with no obstructions (light posts, traffic control boxes, utility poles, etc.); and slopes of 2% or less at all times, including driveway aprons.  In commercial areas, at least 8 feet wide with additional space for a “shy zone” on the sides and for street furniture.  (Code minimums are a width of 3 feet, not including curbs, trees, poles or other obstructions with a 5’x5′ passing zone every 200 linear feet.)  Sidewalks (and the rest of the road) should also be configured to ensure safe and efficient bus, trolley, and train access.
  • Parking – if present at all, 7 to 8 feet; using a “reverse-angle back-in” configuration where possible for greater safety when parkers open doors or pull out.
  • Bike Lanes – 5 to 7 feet wide next to parked cars, 4 to 5 feet wide against a curb, or at 3 to 4 foot edge lines against a curb if there isn’t room for a bike lane; “protected bike lane” or cycle track where traffic is heavy or fast, wide enough for people to ride two-abreast but separated from moving traffic and pedestrians by a curb or the parking lane or planters, etc.
  • Traffic Lanes – 9 to 11 feet wide, depending on traffic volume and nature; on heavily trafficked multiple lane roads, widths should be from 10 to 11.5 feet if felt necessary (perhaps making the center lanes wider than the outside lanes) and if, after serving all the other needs, space permits.


The research quotes were culled from the following, which include additional information as well as other citations:

“Ten-Foot Lanes for Boston: Serving Cyclists and Motorists,” by Christopher Longenbaker, paper for Professor Peter Furth, Northeastern University

“Some Research On Lane Widths,” compiled by Marius Navazo, Research Fellow for LivableStreets Alliance, 2009

“The Influence of Lane Widths on Safety and Capacity: A Summary of the Latest Findings,” by Theordore Petritsch, Sprinkle Consulting

Association of Pedestrian and Bicycle Professionals (

Take our website survey!