Article Text
PDF
Abstract
Evidence suggests that 20 mph zones are an effective intervention to reduce casualties from road traffic crashes in urban areas. This analysis compares the costs of construction of the 20 mph zone intervention in high and low casualty areas in London to the value of casualties avoided over 5 and 10 year time horizons. Probabilistic sensitivity analyses were conducted to quantify uncertainty in the results associated with model parameters. Results indicate a net present value (NPV) of £18 947 (90% credible limits −£75 252 to £82 021 2005 prices) after 5 years and £67 306 (£−29 157 to £137 890) after 10 years when 20 mph zones are implemented in areas with one or more casualty per kilometre of road. Simulations from our model suggest that the ‘threshold of casualties’ where NPVs become positive using a 10 year time horizon is 0.7 casualties per kilometre.
Statistics from Altmetric.com
Background
Area-wide traffic calming schemes aim to discourage non-residential traffic passing through residential areas and have been shown to be an effective strategy for reducing road traffic casualties.1 The 20 mph (32 km/h) zone is an area-wide traffic calming strategy which uses road engineering measures, including vertical and horizontal deflections, to physically slow traffic to a maximum of 20 mph. Over the last 20 years, many 20 mph zones have been established across the UK, with over 400 zones in London alone. Recent evidence has suggested that 20 mph zones are associated with a 42% reduction in road traffic casualties.2
A number of economic evaluations of interventions to reduce speeds in urban areas have been undertaken previously, for example, the introduction of roundabouts in Sweden,3 and speed cameras in Barcelona.4 To the best of our knowledge, only one economic evaluation has been performed assessing the economic impact of speed zones, where the cost-effectiveness of 30 km/h safety zones in urban areas was compared with that of 60 km/h zones in rural areas in the Netherlands.5 We aimed to perform a cost benefit analysis of introducing 20 mph speed zones to reduce casualties from road traffic collisions in London.
Methods
We conducted a cost benefit analysis of all 20 mph zones implemented in London through 2006 (399 zones). Our analyses focused on one outcome, the value of casualties avoided, and one cost, that of zone construction. Using 5 and 10 year time horizons, we simulated forward-looking estimates of investing in new 20 mph zones on high and low casualty roads in London.
Benefits of 20 mph zones
We derived severity-specific estimates of the effects of 20 mph zones (ie, on fatal, serious and minor injuries) from analyses of geographically-coded police road casualties data (STATS19) for the period 2000 to 2006. We excluded 14 zones from analysis because of a lack of information on their implementation dates. We used conditional fixed effects Poisson models to examine the change in casualty counts on roads inside and outside 20 mph zones allowing for London-wide trends in casualties (table 1). These methods are described in detail elsewhere.2 Analyses were restricted to data from 2000 to 2006 for two reasons; (1) to better reflect the impact of implementing a new 20 mph zone, as evidence suggests that the effectiveness of 20 mph zones appears to have decreased in more recent years compared to those implemented 20 years ago2 and; (2) to be more consistent with the cost data described below.
Percentage reduction in casualties in 20 mph zones
Value of benefits
The value of avoided casualties was based on data provided by the Department for Transport (DfT) whose published values include lost output, human costs and medical and ambulance costs.6 The DfT estimate the average value of prevention per casualty at £1 428 180 for fatalities, £160 480 for seriously injured casualties, and £12 370 for slightly injured casualties (in 2005 GBP). We discounted these values at an annual rate of 3.5% as per Her Majesty's (HM) Treasury guidelines.
Costs
We were able to obtain estimates of the costs for 187 (47%) of 399 zones in London. The costs included those associated with their construction and with 20 mph zone consultations. To better reflect the current costs of implementing a new 20 mph zone, we based cost estimates in our analysis on data available for 159 (58%) of the 276 zones completed in, or after, the financial year 2003–2004. We excluded data from five zones which were part of a Home Zone (a more costly intervention whereby residential streets are specifically designed with shared space for vehicles and other road users to reduce vehicle speeds), and from a further 10 zones within which substantial traffic-calming measures had already been implemented. All costs were adjusted for inflation using the Consumer Price Index and are reported in 2005 GPBs. We calculated the cost per kilometre of road for each zone as follows: mean: £59 334.16 per kilometre, SD: £72 985.81.
Uncertainty
In order to quantify uncertainty in the results associated with model parameters, we conducted a probabilistic sensitivity analysis, running 10 000 simulations with parameter estimates randomly sampled from the following statistical distributions: γ distribution for costs; β distribution for effect of 20 mph zones on casualties (table 1), and normal distribution for the annual decline in casualties in all roads in London7 (table 1). We were unable to consider the uncertainty around the monetary values of casualties prevented.
Model
For our simulations we stratified 20 mph zones into two groups based on road injury risk: a ‘low casualty’ group, with >0 and <1 casualty per kilometre of road per year in the 3 years before zone construction, and a ‘high casualty’ group with ≥1 casualty per kilometre of road per year. Table 2 shows the observed 2004–2006 average annual casualties per kilometre in 20 mph zones in low and high casualty areas. We did not simulate the costs and benefits of implementing 20 mph zones in areas of London with no history of road traffic casualties.
Average annual casualties (2004–2006) per kilometre in low and high casualty 20 mph zones
Our model assumes that areas of road can either receive the 20 mph zone treatment or remain untreated. In the first year, we assumed that casualties in areas where roads had received the 20 mph zone intervention declined by the percentage reduction in casualties attributable to 20 mph zones, plus the annual reduction in casualties seen on all roads in London (ie, the background trend in casualty reduction). In subsequent years, we assumed that casualties in treated areas declined according to the background trend only. In other words, we assumed that there would be a step-change at the point of introducing the 20 mph zone. Casualties in areas where roads are not part of a 20 mph zone were assumed to decline according to the background trend each year only.
Results are presented in terms of net present values (NPVs) using 5- and 10-year time horizons, which indicate the sum of the costs and benefits of the zones after discounting. Positive values suggest that the societal benefits are greater than the costs of the zones. 90% credible limits were derived from results of the 10 000 simulations of the probabilistic sensitivity analysis. To obtain a ‘threshold of casualties’ where the benefits of casualty reduction outweigh the costs of construction, we simulated investment in roads at a range of casualty rates and examined the level of casualties where the mean NPV was roughly zero.
Results
Table 3 presents the estimates of the (discounted) benefits and NPVs with 90% credible limits, stratified into low and high casualty roads.
Societal benefits and net present value (NPV) over 5- and 10-year time horizons of 20 mph zones
Taking a time horizon of 5 years, 20 mph zones provide benefits in terms of casualty reductions worth an average of £23 344 per kilometre of road segment in low casualty areas, while the NPV of constructing the 20 mph zones was −£36 117 per kilometre (ie, the cost of implementing the zone was, on average, greater than the value of casualties prevented).
In the high casualty areas, after 5 years the average value of prevented casualties was £78 940 per km of road segment, which is £18 947 more than the average cost of implementation.
Taking a time horizon of 10 years, the 20 mph zones provided benefits in terms of casualty reductions worth an average of £37 278 per km of road in low casualty areas. While the NPV was −£22 183 (ie, average costs of implementation were still greater than the average value of prevented casualties), our simulations suggest that the benefits outweighed the costs in 53% of zones implemented in low casualty areas.
Over a 10-year time horizon, the cost-benefit calculations suggest a base case NPV of £67 306 and positive NPVs for 85% of roads with ≥1 casualty per km per year. Simulations from our model suggest that the ‘threshold of casualties’ where NPVs become positive using a 10 year time horizon is 0.7 casualties per kilometre. This means that, on average, the societal benefits of 20 mph zones will outweigh the costs of construction after 10 years when the intervention is implemented on roads with >0.7 casualties per kilometre.
Discussion
Implementing 20 mph zones in high casualty areas (≥1 casualty per km per year) is a cost-effective way to reduce road injury. However, the costs of implementation in low casualty areas and areas with no casualty history outweigh the benefits.
Limitations of cost-benefit analysis
Casualty reduction is only one reason that local highways authorities choose to implement 20 mph zones. Our analysis did not include other probable benefits of zones, such as increases in walking and cycling, improved community coherence and fewer damage only collisions. Our cost data were similarly limited in scope: no allowance was made for costs of maintenance of zones, or of possibly slower response times of emergency services.
Our analysis might be particularly criticised for not including the costs of longer car journeys as there are methods available to value the travel time lost due to the intervention.8 Unfortunately we could not access data of sufficient quality on pre-intervention average speeds and traffic volumes in London. In London, average travel speeds are very low (less than 10 mph in Central London, 12 mph in Inner London, and between 18–22 mph in Outer London9). Given that the intervention was implemented on mostly minor roads, and travel on minor roads typically makes up a small proportion of any particular journey, the 20 mph zone intervention is perhaps unlikely to substantially increase overall travel times.
A further limitation is that while our probabilistic sensitivity analysis was able to present the implications of parameter uncertainty, it was unable to deal with variability among subgroups; for example, zones located in different types of areas (eg, residential or commercial areas; urban or suburban areas) may require different costs for implementation.
Despite its limitations, our analysis suggests that 20 mph zones are a cost-effective way of reducing casualties in areas with more than an average 0.7 casualties per km per year. Similar to other studies comparing the costs and benefits of other types of speed reduction interventions,3–5 we found that the speed reduction can have favourable economic impact in areas with a previously high frequency of casualties.
What is already known on this subject
-
20 mph zones are an effective way of reducing casualties on urban roads.
What this study adds
-
The benefits of 20 mph zones outweigh the costs of implementation on roads with a previously high frequency of casualties.
Nigeria releases funds to combat lead poisoning
Nigeria pledged $5.4 million to combat lead contamination that has killed at least 400 children. The funds are also intended to treat thousands who have been poisoned by contamination from gold mining. Although promised some while ago, these funds were only recently released. Doctors Without Borders is prepared to treat 1500 children suffering from lead poisoning and to do lead cleanup work before the rainy season begins. The gold rush led to what some say is the worst-ever lead poisoning crisis. Reuters, Bloomberg. Editor's comment: We used to think it was only paint chips and gasoline that were responsible for lead poisoning. It seems a terrible irony that in this case, it is a by-product of gold mining.
Acknowledgments
20 mph zone data were supplied by Peter Sadler at the London Road Safety Unit.
Footnotes
-
Contributors All authors contributed to the design of the study, analysis of the data, interpretation of results and the writing of the manuscript.
-
Funding This work was supported by Transport for London contract No TfL2218. The views expressed in the article are those of the authors and not necessarily those of Transport for London.
-
Competing interests None.
-
Provenance and peer review Not commissioned; externally peer reviewed.