The single biggest factor in road deaths worldwide
Speed sits at the centre of road safety. It influences both the likelihood of a crash happening and how severe the outcome will be when it does. No other risk factor shapes so many deaths across so many countries — and no other risk factor is so consistently underestimated by the public.
This page explains what speeding is, why it kills, what the evidence says, and which interventions actually work.
What counts as speeding?
Speeding takes two forms, and both matter:
- Excessive speed: driving above the legal speed limit.
- Inappropriate speed: driving within the limit but too fast for the conditions — rain, fog, darkness, sharp bends, or the presence of pedestrians and cyclists.
A driver doing 50 km/h past a school at drop-off time may be perfectly legal and dangerously fast at the same time. Road safety policy has to address both.
Why speed kills: the physics
The relationship between speed and harm is not linear — it is exponential. Two mechanisms drive this:
1. Crash likelihood. Higher speeds mean longer stopping distances, less time to react, and a narrower field of effective vision. Research consistently finds that as average traffic speeds rise, crash frequency rises with them — and the effect on fatal crashes is steepest. A widely used rule of thumb from international research: a 1% increase in average speed produces roughly a 4% increase in fatal crash risk.
2. Crash severity. Kinetic energy increases with the square of speed. The human body can only tolerate limited crash forces, and those limits define what road safety professionals call survivable speeds:
| Crash type | Approximate survivable speed |
|---|---|
| Car striking a pedestrian or cyclist | ~30 km/h |
| Side impact between cars (e.g. at intersections) | ~50 km/h |
| Head-on impact between cars on undivided roads | ~70 km/h |
Above these thresholds, the probability of death climbs steeply. A pedestrian struck at 50 km/h is several times more likely to die than one struck at 30 km/h. These thresholds are why the Safe System approach ties speed limits to road function: 30 km/h where cars mix with people walking and cycling, and higher speeds only where the infrastructure physically prevents the deadliest crash types.
The corollary is hopeful: small speed reductions produce outsized safety gains. Cutting average speeds by even a few km/h measurably reduces deaths across a network.
The scale of the problem
The World Health Organization identifies speed as a critical factor in road traffic deaths globally, contributing to a substantial share of the world’s road fatalities — commonly estimated at around one in three deaths in many high-income countries, and a comparable or greater share in many low- and middle-income countries where enforcement capacity is weaker and road environments are more mixed.
Speeding is also strikingly normalised. Roadside surveys across many countries routinely find that a large proportion of drivers exceed posted limits in free-flowing traffic, particularly on urban roads — precisely where pedestrians and cyclists are most exposed.
What works: the evidence-based toolkit
Decades of evaluation have produced a clear picture of effective speed management. The strongest results come from combining measures, not relying on any single one.
1. Safe speed limits
Setting limits according to survivable-speed principles is the foundation. The most significant recent trend is the spread of 30 km/h (20 mph) limits in urban areas. Spain applied 30 km/h as the default on most urban streets nationwide in 2021; Wales moved its default urban limit to 20 mph in 2023; and hundreds of cities worldwide — from Paris and Brussels to Bogotá — have adopted area-wide low limits. Evaluations consistently show reduced speeds, reduced casualties and, contrary to frequent predictions, minimal effects on journey times in dense urban traffic.
2. Self-explaining, self-enforcing roads
Drivers respond more to road design than to signs. Narrowed lanes, raised crossings, chicanes, roundabouts and street trees all signal — and physically encourage — lower speeds. Engineering that makes the safe speed feel like the natural speed is more durable than enforcement alone.
3. Automated enforcement
Speed cameras are among the most thoroughly evaluated interventions in road safety. Systematic reviews find they reliably reduce speeding, crashes, and deaths and serious injuries at treated locations. Average-speed (point-to-point) cameras, which measure speed over a stretch of road rather than at a single point, produce especially consistent compliance and are increasingly used on major routes.
For enforcement to retain public legitimacy, revenue transparency matters: jurisdictions that ring-fence camera revenue for road safety investment tend to sustain stronger public support.
4. Intelligent Speed Assistance (ISA)
ISA uses digital maps and camera-based sign recognition to inform, warn, or actively discourage drivers from exceeding the limit. Since July 2024, all new cars sold in the European Union must be fitted with ISA. It is the most significant vehicle-based speed measure yet deployed at scale, and its real-world effects will be a defining road safety story of this decade.
5. Penalties that change behaviour
Fines alone have limits, particularly where they fall unevenly across incomes. Demerit/points systems, licence suspension for gross violations, and — in several countries — income-scaled “day fines” (as used in Finland and Switzerland) aim to make deterrence more consistent. Evidence favours certainty of detection over severity of punishment: drivers respond more to a high chance of being caught than to the size of a rare penalty.
6. Public engagement
Communication campaigns work best when they support visible enforcement and explain why limits are set as they are — the survivable-speed logic — rather than simply exhorting drivers to slow down.
Common myths, briefly answered
“Speed limits are about revenue, not safety.” The physics of crash severity is not a revenue mechanism. Jurisdictions can and should be transparent about enforcement revenue — but the survivable-speed evidence stands independently of who collects the fines.
“Faster roads are fine if you’re a good driver.” Speed reduces every driver’s margin for error and increases everyone’s stopping distance — including the crash-free driver who meets someone else’s mistake. The Safe System approach exists precisely because human error is universal.
“Lower limits just cause congestion.” In urban traffic, average journey speeds are governed mainly by junctions and traffic volume, not by the limit between them. Evaluations of 30 km/h cities have found journey-time effects to be small.
“Only extreme speeding matters.” Because risk rises continuously with speed, the large number of drivers travelling slightly above safe speeds contributes substantially to total deaths — not just the small number travelling far above them.
The bottom line
Speed is the risk factor that connects all others: it magnifies the consequences of every lapse in attention, every misjudgement, every drink taken before driving. It is also the risk factor with the most complete evidence base for action. Countries that have brought average speeds down — through survivable-speed limits, self-enforcing road design, automated enforcement and vehicle technology — have saved lives at a scale that few other public health interventions can match.
Sources and further reading: World Health Organization — Global Status Report on Road Safety and WHO speed management guidance; International Transport Forum (ITF/OECD) — Speed and Crash Risk; European Transport Safety Council (ETSC) — publications on ISA and enforcement; iRAP — Speed and star rating resources. Figures cited are drawn from this body of international research; see each organisation’s latest publications for current data.
Part of the Road Safety News Risk Factors series: Speeding · Drink & drug driving · Seat belts & child restraints · Helmets · Distracted driving · Fatigue.
