1.1 Speed Limits

Speed limits are only one part of the system that attempts to control driving speeds. Well-established speed limits based on the use of appropriate engineering practices form the basis for roadway design and operations. Active enforcement and supportive adjudication are also essential to support established limits (NHTSA, FHWA, & FMCSA, 2014).

Speed limits are set both by legislation and by administrative action. General speed limits apply to all roads in a class, such as rural interstates or local streets. They are set by State, municipal, or even at times by Federal law based on tradeoffs between safety, travel efficiency, and community concerns, taking into account the design characteristics of each road class. Speed zones apply to road segments where the general speed limit is thought to be inappropriate. Speed limits in these zones usually are set by administrative action based on the road segment’s free- flowing travel speeds, crash experience, road and land use conditions, and other factors (TRB, 1998).

Use: A speed limit is in effect on all road segments in all States. For summaries of each State’s maximum speed limits see the Governors Highway Safety Association (2018a) and the Insurance Institute for Highway Safety (2019a) websites. NHTSA (2013) provides an updated publication with each State’s complete speed limit laws.

Effectiveness: The effects of maximum speed limits on speeds, crashes, and casualties have been studied extensively over the past 40 years. However, recent actions by States raising maximum limits, as well as changes in road design, hardware, vehicles, and drivers suggest that new studies may be needed. In 1974 the 55 mph National Maximum Speed Limit (NMSL) was enacted to conserve fuel. Travel decreased, speeds decreased on roads where the speed limit was lowered to 55 mph, and total traffic fatalities decreased by 9,100 from 1973. The slower and more uniform speeds due to the 55-mph limit are judged to have saved between 3,000 and 5,000 lives in 1974 (TRB, 1984). As fuel became plentiful again, travel increased and compliance with the 55-mph limit decreased markedly. In 1987 Congress allowed States to raise speed limits to 65 mph on rural interstate highways. States that raised their limits generally saw increases of about 4 mph in average speeds and 85th percentile speeds and statistically significant increases in traffic fatalities on these roads. In 1995 Congress repealed the NMSL and returned full authority to set speed limits back to the States. Again, increased speed limits produced modest increases in both average and 85th percentile speeds as well as increases in traffic fatalities (TRB, 1998; TRB, 2006). Speed limit increases from 75 to 80 mph on rural Texas interstates in 2006 also resulted in increased speeds relative to a comparison highway where the limit wasn’t changed (Retting & Cheung, 2008). Utah increased maximum speed limits on certain highways from 75 to 80 mph in 2010 and again in 2013. After the limit increases, average vehicle speed increased by about 3 mph, and vehicles were significantly more likely to surpass 80 mph (Hu, 2017). A recent study found that each 5 mph increase of the State maximum speed limit was associated with an 8% increase in fatality rates on interstates and freeways and a 4% increase on other roads (Farmer, 2016). The study estimated that there were 33,000 more traffic fatalities between 1995 and 2013 than would have been expected if State maximum speed limits had not increased since 1993.

Lower maximum speed limits definitely reduce crashes and casualties when lower limits result in reduced speeds. In general, speeds tend to decrease, but to a lower degree than the reduction in limits. Similarly, when limits are raised, speeds tend to increase by a smaller amount than the change in limits. The same holds true on any road: if a lower speed limit yields reduced operating speeds, crashes and injuries are expected to decrease (AASHTO, 2010). A more comprehensive effort that includes changes to the roadway and/or enhanced enforcement may be required to reduce travel speeds by the desired amount, especially if the road design does not reflect the desired speed limit and operating speeds (TRB, 1998). The State of Victoria, Australia, implemented a comprehensive effort to reduce speeds that combined review and adjustment of speed limits, covert and overt forms of enforcement, a media campaign, penalty restructuring, and other efforts. An evaluation found these combined elements reduced injury crashes by 10% and fatal crashes by 27% (D’Elia et al., 2007).

Several studies examined the effects of speed limit changes (both increases and decreases) in Hungary. For example, urban speed limits were decreased from 60 km/h to 50 km/h (approximately 37 mph to 31 mph) in March 1993, and this change resulted in a decrease in mean speed by 8% and a decrease in road fatalities by 18%. In other cases rural speed limits were increased from 80 km/h (50 mph) to 90 km/h (56 mph) in May 2001, and the mean speed increased by 2.5% and fatalities increased by 13%. A detailed description and more case studies (Australia, Denmark, Norway, etc.) are provided in Speed and Crash Risk (International Transport Forum, 2018).

When urban speed limits were increased from 50 to 70 km/h (from 31 to 43 mph) or from 70 to 80 km/h (from 43 to 50 mph) on 19 urban road segments in Hong Kong, crashes increased by 20 to 30% (Wong et al., 2005).

A systematic evaluation of changed speed limits on rural roads and motorways in Sweden also found fairly consistent increases in travel speeds on all types of rural roads when limits were raised and decreases on roads where limits were lowered (TRB, 1998). Increases of the posted speed limit by 10 km/h (6.2 mph) led to increases in speeds on the order of about 3 to 3.6 km/h (1.9 to 2.2 mph) in mean speeds (weighted for segments length and volume, and including all vehicles on a section for a given time period, not just free flow speeds). Decreases of the posted speed limit of 10 km/h (6.2 mph) led to decreases of about 2 to 3.3 km/h (1.2 to 2 mph) for most road types (Vadeby & Forsman, 2014). These findings are generally in line with those of earlier studies of the effects of changing limits by 5 or more mph.

Relatively few studies have examined the safety effects of speed limit changes on lower-speed roads. Earlier studies found little effect on driving speeds or crash rates when speed limits were raised to near the 85th percentile travel speed or lowered to near the 35th percentile speed, either on rural roads or on urban and suburban arterials (TRB, 1998, p. 6). However, a study from Edmonton, Alberta, Canada, found that speeds on residential streets decreased significantly when limits were lowered and supported with enforcement or other measures (Islam et al., 2013). Specifically, this study found significant speed reductions (3.9 to 4.9 km/h [2.4 to 3.0 mph], 3 and 6 months after treatment, respectively) when posted speed limits in residential areas were reduced from 50 km/h (31 mph) to 40 km/h (25 mph). Changes in posted limits were accompanied by education and enforcement measures, but no changes were made to the roadway. Speeds were reduced on both collector and local road types, in all types of communities, for light and heavy vehicles, for different times of day and on weekends and weekdays. Compliance improved over time up to 6 months post-implementation. Following the lowering of urban default maximum speed limits from 60 km/h (37.3 mph) to 50 km/h (31.1 mph) in 2003 in Adelaide (South Australia), low speed roads showed a significant reduction in mean speed from 46.9 km/h (29.1 mph) to 44.8 km/h (27.8 mph) (Kloeden & Woolley, 2012). From 2003 to 2010 yearly mean speeds have remained lower than before the limits were changed, fluctuating between a high of 44.8 km/h (27.8 mph) and a low of 43.3 km/h (26.9 mph). A follow-up study (Kloeden & Wooley, 2017) found a general downward trend in speeds from 2003 to 2016. In 2016 the mean speed was 41.67 km/h (25.9 mph).

Costs: The immediate costs of changing speed limits are for new signage and for publicizing the new limit. Enforcing the new limit may involve substantial costs.

Time to implement: Speed limit changes can be implemented quickly, as soon as signage is in place and the new limits are publicized.

Other issues:

• Public acceptance, roadway characteristics, enforcement, and publicity: Lowering speed limits can reduce average driving speeds, but it is generally difficult to enforce and obtain broad compliance with a lower speed limit on a roadway designed for much higher speeds (TRB, 1998). Thus, speed limits must be considered as part of a system including roadway design and other characteristics, active enforcement, and publicity.
• Rational speed limits: Speed limits on many road segments are frequently not obeyed, and average travel speeds on these segments substantially exceed the speed limit. One strategy that has been proposed to increase overall safety is to carefully set and enforce credible speed limits for homogeneous road segments. Once credible, also called rational, speed limits are established, aggressive enforcement is used to enforce close to the actual limit. The goal of this strategy is to increase the public’s overall acceptance of speed limits while reducing the number of people driving at speeds considerably higher than the limit. Evidence suggests that drivers’ perceptions of safe speed are in fact influenced by their expectation of what speed above the limit would trigger a ticket (Mannering, 2009). Therefore, lower tolerances would help to increase the perception of the risk of exceeding limits by even small amounts. Although consistency in speed limit setting practices should provide better information about appropriate speeds to drivers, the safety effects of combining rational speed limit setting (with limits raised to between the 50th and 85th percentile free flow operating speed) with enhanced enforcement close to the new limit are uncertain. Reviews of the evidence suggest that it can be difficult to implement or sustain enhanced levels of enforcement. In general, higher speed limits are very likely to lead to higher average speeds if nothing is done to the road or enhanced enforcement is not maintained (Hauer, 2009). Higher average speeds are predicted to lead to increases in fatal and injury crashes (ASHTO, 2010). When testing the effects of raising speed limits, followed by enhanced enforcement in Mississippi and Virginia, average speeds increased in both locations.

In Virginia average speeds tended to increase about 2 mph at locations where the limit was raised by 5 mph and by 3 to 4 mph where it was raised by 15 mph (Freedman et al., 2007). Average speed in Virginia increased by a statistically significant 3 to 4 mph when the limit was raised from 55 to 65 mph on two rural highways (Fontain et al., 2007). Speed variance did not increase and compliance overall was improved in Virginia, which supplemented stricter enforcement with enhanced roadside signs, media publicity, and brochures. Average speeds as well as speed variance increased in Mississippi, where limits were increased on different sections of one route by 5 to 15 mph and the number of extreme speeders were not reduced, except on sections where limits were increased by 15 mph (Freedman et al., 2007).

Mississippi chose to enforce only flagrant violators (at least 5 mph above the limit). Crash effects were inconclusive over both of these fairly short-term evaluations (1 to 1.5 years), although crashes were higher during the Mississippi trial compared to a prior three-year period. A test in Minnesota yielded more promising, though inconclusive crash trends (Harder & Bloomfield, 2007). The Minnesota campaign, which used speeding and crash histories to help target enforcement, effectively reduced mean speeds and especially excessive speeding (speeds of 70 mph and more), but the study period was insufficient to assess crash trends. Extensive radio publicity supplemented by earned media was used in the Minnesota campaign, but it was unclear if these efforts were successful at reaching the target audience.

• Variable speed limits: Speed limits that may adjust to adverse or changing environmental conditions are considered by FHWA to have promise in restoring credibility of speed limits on some highways. Variable speed limits (VSLs) have long been used on European freeways to manage speed and traffic flows. As of 2013 five metropolitan areas in the United States are employing enforceable, variable speed limits on freeways (posted on changeable message signs) (Office of the Assistant Secretary for Research and Technology, 2013). Variable speed limits are also being used in work zones by 11 agencies including Utah DOT, Minnesota DOT, and Texas DOT (Office of the Assistant Secretary for Research and Technology, 2013; National Work Zone Safety Information Clearinghouse, 2016). A high-quality study of safety effects of variable limits deployed on freeways in the St. Louis area reported crash reductions of 8%. The congestion relief benefits were not as high as the public and agencies had hoped, however, leading to somewhat equivocal support for the measure (Bham et al., 2010). No other quality evaluations are available at present. Preliminary investigation of a Wyoming freeway VSL system showed speed reductions from 0.47 to 0.75 mph for every mph reduction in speed limit (Buddemeyer et al., 2010). Other States that have used VSL systems to alter speed limits for weather conditions include Alabama, Delaware, and Washington (Katz et al., 2012). Automated speed enforcement could potentially be linked to variable limits to increase compliance.
• Work zone speed limits: If drivers perceive that limits are too low, workers are not present, and other changes to the roadway do not seem to justify the lower limits, they may not comply, and extensive enforcement may be needed to enforce the limit (Ullman et al., 2013). Sharma et al. (2017) collected data from nine construction work zones in Iowa during 2014 and 2015. The study found consistent speed reductions associated with work zone speed limits, compared to data collected from the time period when work zones were not in place at the same locations.
• Speed limit reductions with advance warning flasher (AWF): In Nebraska seven high-speed intersections equipped with AWFs were selected and tested to examine effects of speed limit reductions in transitional speed zones (Wang & Sharma, 2017). The results showed that a 10-mph reduction from 65 mph posted speed zones led to a 3.8 mph reduction in mean speed, but a 5-mph reduction from a 60 mph posted speed zones did not yield any significant reductions in mean speed.
• Setting speed limits at high-risk locations: A project by Jurewicz et al. (2014) provided guidelines for setting speed limits at high-risk locations in Australia based on the road category/function and the presence of a severe crash risk (i.e., severe crash rate per 100 million vehicle kilometers traveled), types of road use and users, road features, and speeds. The recommendations incorporate other considerations that affect crash risk such as the presences of high-numbers of pedestrians and cyclists, access point density, AADT, among other factors.
• Vision Zero speed limit resolutions: An increasing number of cities are adopting the objectives of Vision Zero to prevent reckless driving, increase safety for all road users, and mitigate road trauma. A range of measures can be used to achieve objectives through speed limit reductions, automated enforcement of speed violations using an expansive network of speed cameras, media campaigns, and engineering measures such as speed humps. New York City is one of the early adopters of the program and enacted a law to implement city-wide speed limits of 25 mph in October 2014 (a decrease from the previous 30 mph) (New York City Mayor’s Office of Operations, 2015). This speed limit reduction potentially halves the fatality risk for a struck pedestrian. Similarly, the Seattle lowered its speed limit to 20 mph on residential streets and to 25 mph on arterials in 2016 (Seattle Department of Transportation, 2017).

---