=60 189 12246 12435 21540 33975 % in 60 zone 59.3% 81.3% 81.0% 79.4% 79.9% Table 21 shows that, in the metropolitan area, 79% of non-casualty crashes on roads zoned 60 km/h and over occurred on roads zoned 60 km/h. Some of the 60 km/h roads would have been collectors, rather than arterial roads. In addition, it is possible that the 60 km/h arterial roads had somewhat higher crash rates than other arterial roads and so using crashes to estimate volumes may lead to overestimation. For these reasons, it was decided to assume that 65% of travel on arterial roads in the metropolitan area occurs on sections zoned 60 km/h. The 65% proportion was applied to the urban arterial travel estimates in Austroads (2000) to estimate the amount of travel on 60 km/h arterial roads. Link length The length of each affected road type also needs to be estimated for input to the MASTER framework. The estimates of lengths of various classes of urban roads are summarised in Table 22. These estimates were derived from published data (Austroads, 2000) and other assumptions as summarised below. For the purpose of this calculation, it was assumed that 80% of the length of urban local roads is composed of local streets and the remaining 20% comprise collector roads. The length of local streets zoned 50 km/h was estimated as the entire length of local streets in Victoria and the ACT and 90% of the length of local streets in NSW and Queensland. It was assumed that 60% of the length of urban arterial roads is zoned 60 km/h. Applying this percentage to the estimates of total length of arterial roads in Austroads (2000), produces the estimates summarised in Table 22. Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 41 Table 22. Estimated length of urban roads in Australia in 1999 (kms). Estimates for urban local roads and urban arterial roads are from Austroads (2000). It was assumed that 80% of the length of urban local roads is composed of local streets and the remaining 20% is urban collector roads. The length of local streets zoned 50 km/h is estimated as the entire length of local streets in Victoria and the ACT and 90% of the length of local streets in NSW and Queensland. Road type NSW Vic Qld WA SA Tas NT ACT Total Urban local 21,103 26,400 14,814 10,469 7,431 2,728 59 1,830 84,834 Local streets 16,882 21,120 11,851 8,375 5,945 2,182 47 1,464 67,867 Local streets zoned 50 km/h 15,194 21,120 10,666 0 0 0 0 1,464 48,444 Local streets zoned 60 km/h 1,688 0 1,185 8,375 5,945 2,182 47 0 19,423 Collector 4,221 5,280 2,963 2,094 1,486 546 12 366 16,967 Urban arterial 4,181 3,180 1,524 1,588 929 369 160 510 12,441 Urban 60 arterial 2,509 1,908 914 953 557 221 96 306 7,465 Daily travel was calculated using the formula: Estimated AADT = amount of travel / (length x 365) The estimates of total travel, length and AADT on the different types of urban roads are summarised in Table 23. The estimated AADTs for urban local streets from the current analysis are similar to those in the Victorian RIS (VicRoads, 2000). The ratio of length of local streets to collector roads is much greater in the Victorian RIS, however. For the MASTER spreadsheet the crucial value is total travel, therefore the values of length and AADT are of less interest. In terms of total travel, the current estimates for urban local streets across Australia are about three times the estimates for Victoria. Given that the Victorian total travel estimate was taken from NRTC (1996) which is almost double that found in Austroads (2000), the estimate seems reasonable. Page 42 Evaluation of a 50 km/h Default Urban Speed Limit for Australia Table 23. Summary of estimated values related to amount of travel on urban roads. Type of roads Total travel (million vehicle kilometres) Length (kilometres) AADT (vehicles per day) Urban residential throughout Australia 17,469 67,867 705 Urban residential zoned 50 km/h 11,337 48,444 644 Urban residential zoned 60 km/h 6,132 19,423 865 Urban collector roads 13,725 16,967 2,216 Urban 60 km/h arterial roads 43,897 7,465 16,110 5.1.3 Crash numbers The MASTER spreadsheet requires Australia-wide numbers of injury crashes in urban areas disaggregated by road type (local streets, collectors and arterial roads) and by speed zone. This information was not available. Therefore, crash numbers were estimated by pro-rating available data. It should be noted that the available data were based on reported crashes. This provides a conservative estimate because there is significant under-reporting of non-fatal crashes (particularly non-hospitalisation casualty crashes and property damage only crashes). The Victorian RIS (VicRoads, 2000) estimated that 2,000 casualty crashes occurred each year on urban local streets. Based on this value, it was estimated that approximately 7,000 casualty crashes occur each year on urban local streets throughout Australia. If the crash rates per vehicle kilometre travelled are similar on urban local streets and collector roads, then about 1,570 casualty crashes would have occurred on collector roads in Victoria (based on relative amount of travel on urban local streets and collectors). This would correspond to about 5,500 casualty crashes on urban collector roads in Australia each year. The number of crashes on urban local streets zoned 60 km/h was estimated by pro-rating the amounts of travel. If there are 7,000 casualty crashes per year on all urban local streets throughout Australia and 35% of the travel on urban local streets is on those currently zoned 60 km/h (from travel data), then it is estimated that 2,450 (7,000 x 0.35) casualty crashes occur on urban local streets that are currently zoned 60 km/h. In 1999 (after the introduction of the 50 km/h speed limit for local streets) there were 10,068 casualty crashes on 60 km/h roads in metropolitan areas of NSW (RTA, 2000c). It was assumed that most of these crashes occurred on 60 km/h arterial roads with a smaller number on collector roads. Based on this value, it was estimated that about 23,000 casualty crashes occur on urban arterial roads zoned 60 km/h each year throughout Australia. Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 43 5.1.4 Speed-crash relationship The MASTER spreadsheet allows the form of the speed-crash relationship to be specified. In the analyses reported here, the Andersson and Nilsson (1997) relationship between changes in mean speed and number of crashes was used: nA = (vA/vB)2 * nB where nA = number of injury crashes after speed change nB = number of injury crashes before speed change vA = mean speed after speed change vB = mean speed before speed change This relationship was chosen in preference to the relationship developed by Kloeden et al (1997) because Cameron (2000) found that the risk estimates from Kloeden et al’s relationship were not sufficiently stable for speeds below 60 km/h. The Andersson and Nilsson (1997) relationship between changes in mean speed and crash costs was also used: CA = [k*((vA/vB)2-1)+1]*CB Where CA = crashes costs after speed change CB = crashes costs before speed change vA = mean speed after speed change vB = mean speed before speed change k = a constant depending on the actual unit costs of fatal, serious and minor injuries and the average number of each in casualty crashes of various severities. A value of k=2 was used in the analyses since Kallberg and Toivanen found that this applied in most European countries There is an apparent inconsistency between the changes in mean speed and crashes observed to result from the 50 km/h residential speed limit in NSW and those predicted by the Andersson and Nilsson relationship. In NSW, the measured reduction in mean speeds was of the order of about 1 km/h (see Section 4.1.1). From this speed reduction, the Andersson and Nilsson relationship would predict a reduction in crashes of the order of 3%. However, the before and after analysis showed that 21% fewer crashes occurred than would have been expected if the trend from before treatment had continued. This has led to concerns being expressed about the applicability of the Andersson and Nilsson relationship to urban speeds. An alternative approach that could have been used in this report would have involved substituting the observed NSW crash and speed reductions from NSW into a modified version of the MASTER spreadsheet, rather than using the Andersson and Nilsson relationship. This approach was not taken for several reasons. Firstly, the reported speed reductions in Queensland were considerably larger than reported in NSW. This suggested that the size of the speed reduction might depend on the method of implementation. It is possible that the greater emphasis on enforcement in Queensland Page 44 Evaluation of a 50 km/h Default Urban Speed Limit for Australia compared to NSW (and possibly introduction Southeast Queensland-wide, rather than Local Government Area by Local Government Area) might have resulted in larger speed reductions. Secondly, it was unclear what aspects of speed were actually measured in NSW. The MASTER framework requires cruise and average journey speeds as input. It is unclear whether the NSW measurements reflected cruise speeds, average journey speeds or something in between. Thirdly, the NSW results were restricted to local residential streets. It was unclear whether the speed reductions reported in NSW would generalise to urban collector roads and urban arterial roads currently zoned 60 km/h. The overall outcome of the approach used in this report is likely to be more conservative than the alternative approach. The observed casualty reductions in NSW lie between the outcomes predicted using the Andersson and Nilsson relationship for the 5 km/h and 10 km/h cruise speed reduction scenarios in this report. However, if the reported speed reductions in NSW actually represent cruise speed reductions, then the reductions in average journey speeds are likely to be smaller than in the 5 km/h and 10 km/h scenarios. Thus the costs associated with increases in travel time and vehicle operating costs would be smaller using the alternative approach. 5.1.5 Costs of travel time and crashes Austroads provides estimates of cost of travel time for private and business travel by car and for other vehicle types (Thoresen, 2000). The values for business travel by car and for travel by other vehicle types are considerably higher than those for private travel by car. The Bureau of Transport Economics has published estimates of the cost of crashes at varying levels of severity (BTE, 2000). Based on these figures, Cameron (2000) estimated that the cost of casualty crash was $152,270. However, a fundamental problem exists in comparing these two forms of costs. The BTE crash costs are based on a Human Capital approach in which time lost as a result of crashes is only valued if it is paid work time or “productive” time devoted to unpaid community contributions (child care, housework, voluntary work etc). Leisure time lost through crashes is not valued in this approach. In the Austroads figures, the estimated values of travel time for private use of cars (unpaid time) are lower than those for business use of cars (paid time). However, the unpaid time is assigned a value that is a high proportion of hourly average weekly earnings. Thus, unpaid time is given a value in the travel time estimates, but not in the crash cost estimates. The outcome of this discrepancy is to value time lost as a result of lower travel speeds at a higher rate than time lost because of crashes. This discrepancy is not unique to the Australian estimates and has been discussed at length in the safety literature. Hauer (1994) pointed out that the discrepancy implies that it is better to be dead than stuck in traffic. Miller (1993) warned of the danger of making decisions based on conflicting travel time and crash cost values. He concluded that “by using monetary crash costs in resource allocation, highway engineers inadvertently created mobility by sacrificing lives” (p.605). This report takes two approaches to addressing the discrepancy in the values of time generated by the travel time and crash cost estimates. The first approach attempts to deal with the discrepancy by reducing the estimates of travel times to remove the value of unpaid time (and comparing the adjusted estimates with published values of crash costs). The second Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 45 approach uses published values for travel times but increases the value of crash costs to a level which attempts to include the value of unpaid time. The values of a casualty crash used in the analyses were the BTE (2000) based value of $152,270, a lower estimate of $110,000 and a higher estimate of $250,000. The values of the cost of a non-injury crash used in the current analyses were a BTE (2000) based value of $6,000, a lower value of $4,500 and a higher value of $10,000. 5.1.6 Share of traffic by trip purpose The MASTER framework requires that the percent of trips that are business, private business/commuting and leisure be entered. It also requires that the value of travel time associated with these three trip purposes be entered. The complicating issue in calculating the values to enter is that both of these measures are dependent on vehicle type and road type. The Survey of Motor Vehicle Use 1999 (Table 8) provides information about the percent of travel by different vehicle types (for all types of roads) classified into business use, travel to and from work and “personal and other”. The Mass Limits Review – Road and Bridge Statistical Tables (NRTC, 1996) provides information about the amount of travel on different road types by different vehicle types. Table 24 combines these sources of information to estimate the proportion of trips on urban roads that are for different purposes. The calculations in the Table assume that the proportion of travel (for a given vehicle type) according to trip purpose is the same for all types of roads (e.g. if 20% of car trips are for business, then this is true for each road type). If the percentage of travel on urban local streets that is by private car is greater than for all roads as a whole, then the effect of the assumption would be to overestimate travel time costs. Based on data in Table 8 of the Survey of Motor Vehicle Use 1999 (ABS, 2000), the calculations in Table 24 assume that: • 25% of passenger car and motorcycle travel is for business; • 50% of passenger car and motorcycle travel is for personal business and commuting; • 25% of passenger car and motorcycle travel is for leisure; • 70% of light commercial travel is for business and remaining travel is divided between personal business and leisure; • all rigid and articulated trucks travel is for business. The share of traffic by trip purpose and vehicle type on urban collector roads was assumed to be the same as on urban local streets. Page 46 Evaluation of a 50 km/h Default Urban Speed Limit for Australia Table 24. Travel by trip purpose on urban local streets and urban arterial roads. Based on data from Mass Limits Review – Road and Bridge Statistical Tables (NRTC, 1996) and Table 8 of the Survey of Motor Vehicle Use 1999 (ABS, 2000). Type of travel Urban local streets Urban arterials zoned 60 km/h Distance travelled (million vehicle-kms) % of travel Distance travelled (million vehicle-kms) % of travel Business trips 6,179 33.6 23,971 37.7 Personal business and commuting trips 8,012 43.6 25,962 40.9 Leisure trips 4,187 22.8 13,568 21.4 Travel time costs were estimated by applying the values in Thoresen (2000) (Table 9) to the percentages of travel by each vehicle type and purpose in Table 24. The value of travel time for each type of trip is summarised in Table 25. These values will be used in Approach 2. In Approach 1 the value of travel time for personal business/commuting and leisure trips will be set to zero (as explained earlier). Table 25. Values of travel time used in Approach 1 and Approach 2. Type of travel Urban local streets Urban arterials zoned 60 km/h Approach 1 Approach 2 Approach 1 Approach 2 Business travel $29.77 $29.77 $30.05 $30.05 Personal business and commuting travel $0.00 $12.18 $0.00 $12.18 Leisure travel $0.00 $12.18 $0.00 $12.18 5.1.7 Vehicle operating costs Vehicle operating costs for each average speed were estimated by applying the values in Thoresen (2000) (Table 12) for the Urban Stop-Start model to the percentages of travel by each vehicle type and purpose in Table 24. The used car value was used for private travel by car. The resultant values are summarised in Table 26. Note that vehicle operating costs are greater at lower speed in the Urban Stop-Start model because speed is a component of the denominator of the function. Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 47 Table 26. Estimates of vehicle operating costs ($/km), based on vehicle mix and cruise speeds. Cruise speed level Urban local streets Urban arterials zoned 60 km/h 57 km/h (60 km/h zone) 0.262 0.269 52 km/h (5 km/h reduction) 0.264 0.270 47 km/h (10 km/h reduction) 0.265 0.272 5.1.8 Air pollution values Unit costs for air pollutants emitted by vehicles were taken from Cosgrove (1994). The relationship between amount of air pollutants and travel speed was taken from Ward, Roberston and Allsop (1998). These assumptions were used by Cameron (2000) in his estimation of optimum travel speeds on urban local streets. 5.1.9 Summary of Scenarios The three values of the cost of a casualty crash used were the value used by Cameron (2000) based on BTE (2000), a lower value and a higher value. The three values were used to assess the extent to which the net outcomes were sensitive to the values selected to represent the cost of a casualty crash. The two values of the likely reduction in cruise speed used were 5 km/h and 10 km/h. Eight Scenarios were examined in each analysis. The first six scenarios combine three values of the cost of a casualty crash and two values of the likely reduction in cruise speed associated with a reduction in the speed limit from 60 km/h to 50 km/h. In these six analyses, the value of travel time is adjusted to remove the effects of unpaid time. The final two scenarios use the published value of travel time and the higher value of crash costs for 5 km/h and 10 km/h reductions in cruise speed. The base case is considered to be a reduction in cruise speed of 5 km/h assessed at the BTE (2000) based crash cost values and the adjusted values of travel time. 5.2 Urban local streets 5.2.1 Analyses in hypothetical and current situations As described in the earlier section, some parts of Australia already have 50 km/h speed limits on urban local streets. Therefore, while the hypothetical assessment of the effects of changing from a 60 km/h limit in urban local streets throughout Australia to a 50 km/h limit may be of theoretical interest, it does not really measure the effects of the proposed change in the default urban speed limit. The more relevant analysis assesses the effect of changing to a 50 km/h limit in those areas where the current residential speed limit is 60 km/h. Both sets of analyses are presented here. Analysis 1 estimates the benefits and costs of implementing a default 50 km/h speed limit on urban local streets across Australia, compared to a baseline situation that all local streets are zoned 60 km/h. Page 48 Evaluation of a 50 km/h Default Urban Speed Limit for Australia Analysis 2 estimates the benefits and costs of implementing a default 50 km/h speed limit on urban local streets across Australia, compared to a baseline situation that represents the current situation. The current situation is represented here as 50 km/h on all local streets in Victoria and the ACT and 90% of all local streets in NSW and Queensland. 5.2.2 Analysis 1: Australia-Wide 60 Km/H Versus 50 Km/H This analysis estimates the benefits and costs of implementing a default 50 km/h speed limit on urban local streets across Australia, compared to a baseline situation that all local streets are zoned 60 km/h. Table 27 summarises the estimated outcomes of a default 50 km/h speed limit on urban local streets throughout Australia measured from 60 km/h national baseline. It assumes that the speed limit reduction would result in a 5 km/h reduction in cruise speed. The savings in costs of casualty crashes are more than an order of magnitude greater than the savings in the costs of property damage crashes. The savings in terms of reduced air pollution are relatively modest. The vehicle operating costs are based on the mix of vehicle types on urban local streets. Approach 1 attempts to deal with the discrepancy between published values for crash costs and travel times by reducing the estimates of travel times to remove the willingness-to-pay component. Approach 2 uses published values for travel times but uses the highest value of crash cost savings (which approaches willingness-to-pay estimates). Approach 1 concludes that the outcome is a net benefit (negative value in the Table) if the higher value of crash costs is used. Approach 2 results in an estimate of travel time costs that is almost twice that in Approach 1 and concludes that the outcome is a net loss. If the effect of increased travel time is excluded, all estimates show a net benefit, ranging from $98 million per year to $247 million per year. Table 27. Estimated outcomes of default 50 km/h speed limit on urban local streets throughout Australia (measured from 60 km/h national baseline) if the speed limit reduction results in a 5 km/h reduction in cruise speed. Negative values represent a reduction in costs. All values in $000s per year. Component Approach 1 – Adjusted travel time costs Approach 2 – Austroads travel time costs BTE crash costs lower value of crash costs higher value of crash costs higher value of crash costs Casualty crash costs -149,955 -108,328 -246,200 -246,200 Property damage crash costs -11,760 -8,820 -19,600 -19,600 Air pollution costs -948 -948 -948 -948 Vehicle operating costs 19,518 19,518 19,518 19,518 Net effect (excluding travel time costs) -143,146 -98,579 -247,230 -247,230 Travel time costs 178,922 178,922 178,922 323,261 Net effect (including travel time costs) 35,777 80,344 -68,308 76,031 Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 49 If implementation of a default 50 km/h speed limit on urban local streets throughout Australia (measured from 60 km/h national baseline) resulted in a 10 km/h reduction in cruise speed, the savings associated with fewer casualty and property damage crashes would be more than double that associated with a 5 km/h reduction in cruise speed (see Table 28). Approach 1 concludes that the outcome is a net benefit only if the highest value of crash costs is used. Approach 2 concludes that the outcome is a net loss. If the effect of increased travel time is excluded, all estimates show a net benefit, ranging from $239 million per year to $605 million per year. The costs associated with increased travel times for a 10 km/h reduction in cruise speed are more than double the costs associated with a 5 km/h reduction in cruise speed. The travel time calculations are based on average speed, not cruise speed, and the reduction in average speed for a 10 km/h reduction in cruise speed is more than double that for a 5 km/h reduction in cruise speed. Table 28. Estimated outcomes of default 50 km/h speed limit on urban local streets throughout Australia (measured from 60 km/h national baseline) if the speed limit reduction results in a 10 km/h reduction in cruise speed. Negative values represent a reduction in costs. All values in $000s per year. Component Approach 1 – Adjusted travel time costs Approach 2 – Austroads travel time costs BTE crash costs lower value of crash costs higher value of crash costs higher value of crash costs Casualty crash costs -369,978 -267,273 -607,438 -607,438 Property damage crash costs -28,560 -21,420 -47,600 -47,600 Air pollution costs -3,010 -3,010 -3,010 -3,010 Vehicle operating costs 52,547 52,547 52,547 52,547 Net effect (excluding travel time costs) -349,001 -239,156 -605,501 -605,501 Travel time costs 481,229 481,229 481,229 870,319 Net effect (including travel time costs) 132,228 242,073 -124,272 264,818 While the monetary values associated with the increases in travel times appear very large, these values may be illusory. Travel time increased by 3.7% for a 5 km/h reduction in cruise speed and by 10.0% for a 10 km/h reduction in cruise speed. Expressed in absolute terms, the actual increase in travel times varied from 48,957 hours per day (5 km/h reduction in cruise speed) to 131,808 hours per day (10 km/h reduction). These travel time increases correspond to an average increase in travel time of between 8.8 seconds per day and 23.7 seconds per day for each member of the Australian population. If each person makes four trips per day (on average), then forfeiting 2.2 to 5.9 seconds per trip is required to prevent between 492 and 1,200 casualty crashes per year. The validity of aggregating small changes in travel time will be discussed in Section 5.5.1. Page 50 Evaluation of a 50 km/h Default Urban Speed Limit for Australia 5.2.3 Analysis 2: Current Situation Versus Australia-Wide 50 Km/H This analysis estimates the benefits and costs of implementing a default 50 km/h speed limit on urban local streets across Australia, compared to a baseline situation that represents the current situation. The current situation is represented here as 50 km/h on all local streets in Victoria and the ACT and 90% of all local streets in NSW and Queensland. Table 29 summarises the estimated outcomes of a default 50 km/h speed limit on urban local streets throughout Australia measured from the current baseline. It assumes that the speed limit reduction would result in a 5 km/h reduction in cruise speed. The outcome is a net benefit only for Approach 1 with the highest value of crash costs. For all other Scenarios, the value of increased travel time costs exceeds the saving associated with casualty and property damage crash reductions. Table 29. Estimated outcomes of default 50 km/h speed limit on urban local streets throughout Australia (measured from current baseline) if the speed limit reduction results in a 5 km/h reduction in cruise speed. Negative values represent a reduction in costs. All values in $000s per year. Component Approach 1 – Adjusted travel time costs Approach 2 – Austroads travel time costs BTE crash costs lower value of crash costs higher value of crash costs higher value of crash costs Casualty crash costs -52,484 -37,915 -86,170 -86,170 Property damage crash costs -4,116 -3,087 -6,860 -6,860 Air pollution costs -333 -333 -333 -333 Vehicle operating costs 6,853 6,853 6,853 6,853 Net effect (excluding travel time costs) -50,080 -34,482 -86,510 -86,510 Travel time costs 62,764 62,764 62,764 113,511 Net effect (including travel time costs) 12,684 28,282 -23,746 27,001 Table 30 shows that if implementation of a default 50 km/h speed limit on urban local streets throughout Australia (measured from 60 km/h national baseline) resulted in a 10 km/h reduction in cruise speed, the only Scenario that predicts a net benefit is Approach 1 with the highest value of crash costs. For the other Scenarios, the estimated cost of increased travel time exceeds the savings associated with fewer casualty and property damage crashes. Travel time increased by 3.7% for a 5 km/h reduction in cruise speed and by 10.0% for a 10 km/h reduction in cruise speed. Expressed in absolute terms, the actual increase in travel times varied from 17,191 hours/day (5 km/h reduction in cruise speed) to 46,283 hours/per day (10 km/h reduction). These travel time increases correspond to an average increase in travel time of between 3.1 seconds per day and 8.3 seconds per day for each member of the Australian population. If each person makes four trips per day (on average), then forfeiting Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 51 0.8 to 2.1 seconds per trip is required to prevent between 172 and 425 casualty crashes per year. The validity of aggregating small changes in travel time will be discussed in Section 5.5.1. Table 30. Estimated outcomes of default 50 km/h speed limit on urban local streets throughout Australia (measured from current baseline) if the speed limit reduction results in a 10 km/h reduction in cruise speed. Negative values represent a reduction in costs. All values in $000s per year. Component Approach 1 – Adjusted travel time costs Approach 2 – Austroads travel time costs BTE crash costs lower value of crash costs higher value of crash costs higher value of crash costs Casualty crash costs -129,492 -93,545 -212,603 -212,603 Property damage crash costs -9,996 -7,497 -16,660 -16,660 Air pollution costs -1,057 -1,057 -1,057 -1,057 Vehicle operating costs 18,452 18,452 18,452 18,452 Net effect (excluding travel time costs) -122,094 -83,648 -211,869 -211,869 Travel time costs 168,981 168,981 168,981 305,607 Net effect (including travel time costs) 46,888 85,334 -42,887 93,739 5.2.4 Summary of benefits and costs of implementing a default 50 km/h speed limit on urban local streets The base case is assumed to be a 5 km/h reduction in cruise speed estimated using the BTE crash costs and the adjusted values of travel time costs. The analyses show that this would lead to a net disbenefit of about $36 million per year if implemented throughout Australia. This outcome is sensitive to the value of crash costs: the predicted disbenefit is larger if the lower value of crash costs is used and a net benefit of $68 million per year is predicted if the higher value of crash costs is used. The outcome is also sensitive to the size of the reduction in cruise speed. In the unlikely event that the 50 km/h default speed limit led to a 10 km/h reduction in cruise speed, the outcome would be a net disbenefit (unless the highest value of crash costs was used). Using unadjusted values of travel time (Approach 2) leads to a consistent pattern of net disbenefits. The analyses show that the magnitudes of the outcomes were smaller when measured from the current baseline (i.e. 50 km/h already implemented in parts of NSW, Queensland and Victoria). In all of the scenarios examined, the estimated costs associated with increases in travel time were substantial compared to the estimated savings from fewer casualty and property damage crashes. However, it may not be appropriate to include travel time costs when the individual time differences are so small (less than six seconds per trip). If the effect of increased travel Page 52 Evaluation of a 50 km/h Default Urban Speed Limit for Australia time is excluded, all the Scenarios show a net benefit. The validity of aggregating small changes in travel time will be discussed in Section 5.5.1. As noted earlier, the MASTER framework does not include implementation costs such as signage and public education. The values obtained from the MASTER framework can be interpreted as annual returns in a year that follows implementation. Clearly any implementation costs would need to be discounted across a period of years if these were to be included in the calculations. Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 53 5.3 Urban collector roads Table 31 summarises the estimated outcomes of a default 50 km/h speed limit on urban residential collector roads throughout Australia. It assumes that the speed limit reduction would result in a 5 km/h reduction in cruise speed. The outcome is a net benefit only for Approach 1 and the highest values of crash cost savings. For all other Scenarios, the cost associated with increased travel time exceeds the savings from fewer casualty and property damage crashes. Table 31. Estimated outcomes of default 50 km/h speed limit on urban collector roads throughout Australia if the speed limit reduction results in a 5 km/h reduction in cruise speed. Negative values represent a reduction in costs. All values in $000s per year. Component Approach 1 – Adjusted travel time costs Approach 2 – Austroads travel time costs BTE crash costs lower value of crash costs higher value of crash costs higher value of crash costs Casualty crash costs -117,822 -85,115 -193,443 -193,443 Property damage crash costs -2,322 -1,741 -3,870 -3,870 Air pollution costs -745 -745 -745 -745 Vehicle operating costs 15,337 15,337 15,337 15,337 Net effect (excluding travel time costs) -105,552 -72,264 -182,721 -182,721 Travel time costs 140,461 140,461 140,461 254,027 Net effect (including travel time costs) 34,909 68,197 -42,260 71,306 If implementation of a default 50 km/h speed limit on urban collector roads throughout Australia resulted in a 10 km/h reduction in cruise speed, the outcome is a net benefit only for Approach 1 and the highest values of crash cost savings (see Table 32). For all other scenarios, the cost associated with increased travel time exceeds the savings from fewer casualty and property damage crashes. Travel time increased by 3.7% for a 5 km/h reduction in cruise speed and by 10.0% for a 10 km/h reduction in cruise speed. Expressed in absolute terms, the actual increase in travel times varied from 38,472 hours/day (5 km/h reduction in cruise speed) to 103,578 hours/per day (10 km/h reduction). These travel time increases correspond to an average increase in travel time of between 6.9 seconds per day and 18.6 seconds per day for each member of the Australian population. If each person makes four trips per day (on average), then forfeiting 1.7 to 4.7 seconds per trip is required to prevent between 387 and 955 casualty crashes per year. The validity of aggregating small changes in travel time will be discussed in Section 5.5.1. Page 54 Evaluation of a 50 km/h Default Urban Speed Limit for Australia Table 32. Estimated outcomes of default 50 km/h speed limit on urban collector roads throughout Australia if the speed limit reduction results in a 10 km/h reduction in cruise speed. Negative values represent a reduction in costs. All values in $000s per year. Component Approach 1 – Adjusted travel time costs Approach 2 – Austroads travel time costs BTE crash costs lower value of crash costs higher value of crash costs higher value of crash costs Casualty crash costs -290,697 -210,000 -477,273 -477,273 Property damage crash costs -5,730 -4,298 -9,550 -9,550 Air pollution costs -2,366 -2,366 -2,366 -2,366 Vehicle operating costs 41,293 41,293 41,293 41,293 Net effect (excluding travel time costs) -257,500 -175,371 -447,896 -447,896 Travel time costs 378,163 378,163 378,163 683,920 Net effect (including travel time costs) 120,663 202,792 -69,733 236,024 Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 55 5.4 Urban arterial roads The approach assumes that only arterial roads currently zoned 60 km/h would be subject to the default speed limit of 50 km/h. Table 33 summarises the estimated outcomes of a default 50 km/h speed limit on urban arterial roads currently zoned 60 km/h throughout Australia. It assumes that the speed limit reduction would result in a 5 km/h reduction in cruise speed. The outcome is a net benefit unless the lower value of crash costs is used. The size of the net benefit varies from $49 million to $410 million. Table 33. Estimated outcomes of default 50 km/h speed limit on urban arterial roads currently zoned 60 km/h throughout Australia if the speed limit reduction results in a 5 km/h reduction in cruise speed. Negative values represent a reduction in costs. All values in $000s per year. Component Approach 1 – Adjusted travel time costs Approach 2 – Austroads travel time costs BTE crash costs lower value of crash costs higher value of crash costs higher value of crash costs Casualty crash costs -551,548 -398,439 -905,542 -905,542 Property damage crash costs -10,866 -8,150 -18,110 -18,110 Air pollution costs -2,384 -2,384 -2,384 -2,384 Vehicle operating costs 50,774 50,774 50,774 50,774 Net effect (excluding travel time costs) -514,024 -358,199 -875,262 -875,262 Travel time costs 464,654 464,654 464,654 775,883 Net effect (including travel time costs) -49,370 106,455 -410,608 -99,379 Table 34 shows that if implementation of a default 50 km/h speed limit on urban arterial roads currently zoned 60 km/h throughout Australia resulted in a 10 km/h reduction in cruise speed, the outcome would be a net benefit if the higher value of crash costs was used (regardless of whether published or adjusted values for travel time costs are used). For other Scenarios, the cost of increased travel time exceeds the savings from fewer casualty and property damage crashes. Travel time increased by 4.2% for a 5 km/h reduction in cruise speed and by 12.3% for a 10 km/h reduction in cruise speed. Expressed in absolute terms, the actual increase in travel times varied from 112,370 hours/day (5 km/h reduction in cruise speed) to 329,663 hours/per day (10 km/h reduction). These travel time increases correspond to an average increase in travel time of between 20.2 seconds per day (5 km/h reduction) and 59.3 seconds per day for each member of the Australian population. If each person makes four trips per day (on average), then forfeiting 5.1 to 14.8 seconds per trip is required to prevent between 2,350 and Page 56 Evaluation of a 50 km/h Default Urban Speed Limit for Australia 6,000 casualty crashes per year. The validity of aggregating small changes in travel time will be discussed in Section 5.5.1. Table 34. Estimated outcomes of default 50 km/h speed limit on urban arterial roads currently zoned 60 km/h throughout Australia if the speed limit reduction results in a 10 km/h reduction in cruise speed. Negative values represent a reduction in costs. All values in $000s per year. Component Approach 1 – Adjusted travel time costs Approach 2 – Austroads travel time costs BTE crash costs lower value of crash costs higher value of crash costs higher value of crash costs Casualty crash costs -1,448,424 -1,046,343 -2,378,052 -2,378,052 Property damage crash costs -28,536 -21,402 -47,560 -47,560 Air pollution costs -7,567 -7,567 -7,567 -7,567 Vehicle operating costs 148,954 148,954 148,954 148,954 Net effect (excluding travel time costs) -1,335,573 -926,358 -2,284,225 -2,284,225 Travel time costs 1,363,167 1,363,167 1,363,167 2,276,226 Net effect (including travel time costs) 27,594 436,809 -921,058 -7,999 Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 57 5.5 Overall summary of net effects The base case is considered to be a 5 km/h reduction in cruise speed estimated using the BTE crash costs and the adjusted values of travel time costs. The first column of figures in Table 35 presents the results for the base case. It shows that the outcome would be a net benefit if implemented on urban arterial roads currently zoned 60 km/h. From the baseline condition of 60 km/h on local streets throughout Australia, implementation of a default 50 km/h urban speed limit on local streets, collector roads and arterial roads currently zoned 60 km/h, is predicted to result in a net disbenefit of about $21 million per year. Compared to the baseline of the current state of implementation of 50 km/h speed limits on urban local streets, the overall outcome would be a net benefit of about $1.7 million per year. This outcome is sensitive to the value of crash costs selected: a net disbenefit of $203 million per year is predicted if the lower value of crash costs is used but a net benefit of $476 million per year is predicted if the higher value of crash costs is used. Using unadjusted values of travel time (Approach 2) leads to net disbenefits except for urban arterial roads currently zoned 60 km/h. The outcome is also sensitive to the size of the reduction in cruise speed. In the unlikely event that the 50 km/h default speed limit led to a 10 km/h reduction in cruise speed (see Table 36), the outcome would be a net disbenefit unless the highest value of crash costs was used. Using the highest value of crash costs and adjusted travel time costs, the net benefit is estimated to exceed $1 billion per year. For the lower value of crash costs and the adjusted travel time costs, the disbenefit is estimated at over $700 million per year. Table 35. Estimated net effects of 50 km/h default speed limit on urban roads (local streets, collectors and 60 km/h arterial roads) throughout Australia if the speed limit reduction results in a 5 km/h reduction in cruise speed. Negative values are cost savings. (all values in $000s per year) Affected roads Approach 1 – Adjusted travel time costs Approach 2 – Austroads travel time costs BTE crash costs lower value of crash costs higher value of crash costs higher value of crash costs Urban local streets – from 60 km/h national baseline 35,777 80,344 -68,308 76,031 Urban local streets – from current baseline 12,684 28,282 -23,746 27,001 Urban collector roads 34,909 68,197 -42,260 71,306 Urban arterials currently zoned 60 km/h -49,370 106,455 -410,608 -99,379 Total (from 60 km/h national baseline) 21,316 254,996 -521,176 47,958 Total (from current baseline) -1,777 202,934 -476,614 -1,072 Page 58 Evaluation of a 50 km/h Default Urban Speed Limit for Australia Table 36. Estimated net effects of 50 km/h default speed limit on urban roads (local streets, collectors and 60 km/h arterial roads) throughout Australia if the speed limit reduction results in a 10 km/h reduction in cruise speed. Negative values are cost savings. (all values in $000s per year) Affected roads Approach 1 – Adjusted travel time costs Approach 2 – Austroads travel time costs BTE crash costs lower value of crash costs higher value of crash costs higher value of crash costs Urban local streets – from 60 km/h national baseline 132,228 242,073 -124,272 264,818 Urban local streets – from current baseline 46,888 85,334 -42,887 93,739 Urban collector roads 120,663 202,792 -69,733 236,024 Urban arterials currently zoned 60 km/h 27,594 436,809 -921,058 -7,999 Total (from 60 km/h national baseline) 280,485 881,674 -1,115,063 492,843 Total (from current baseline) 195,145 724,935 -1,033,678 321,764 In terms of the relative benefits on different classes of roads, Tables 35 and 36 also show that most of the benefit is derived through implementation on the urban arterial roads currently zoned 60 km/h. 5.5.1 The validity of aggregating small changes in travel time In all of the Scenarios examined, the estimated costs associated with increases in travel time were substantial compared to the estimated savings from fewer casualty and property damage crashes. While the total costs of the travel time increases are very large, they correspond to very small increases in time for a very large number of trips. Table 37 shows that if implementing a 50 km/h default urban speed limit on local streets, collector roads and arterial roads currently zoned 60 km/h resulted in a 5 km/h reduction in cruise speed, the time increase per trip would be less than 10 seconds. This small increase in travel time would prevent about 3,000 casualty crashes per year. Previous analyses of travel time effects of reduced speed limits have questioned the meaningfulness of valuing very small amounts of travel time across large numbers of vehicles (Austroads, 1996; Hauer, 1994; VicRoads, 2000). Hauer (1994) cites Strand (1993) as supporting the view that it is nonsense to sum the extra few seconds apiece that many vehicle occupants wait at a STOP sign (as compared to a GIVE WAY sign) and compare this value with estimates of crash cost savings. Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 59 Table 37. Summary of increases in travel time costs and travel time increases per trip compared to casualty crash reductions. 5 km/h reduction in cruise speed 10 km/h reduction in cruise speed Cost of time increase ($000) Time increase per trip Casualty crashes saved Cost of time increase ($000) Time increase per trip Casualty crashes saved Adjusted AUST-ROADS (sec) Adjusted AUST-ROADS (sec) Urban local streets – from 60 km/h national baseline 178,922 323,261 2.2 492 481,229 870,319 5.9 1,200 Urban local streets – from current baseline 62,764 113,511 0.8 172 168,981 305,607 2.1 425 Urban collector roads 140,461 254,027 1.7 387 378,163 683,920 4.7 955 Urban arterials currently zoned 60 km/h 464,654 775,883 5.1 2,350 1,363,167 2,276,226 14.8 6,000 Total (from 60 km/h national baseline) 784,037 1,353,171 9.0 3,229 2,222,559 3,830,465 25.4 8,155 Total (from current baseline) 667,879 1,143,421 7.6 2,909 1,910,311 3,265,753 21.6 7,380 The Austroads report on Urban Speed Management in Australia (Austroads, 1996) concludes that: “Economic theory requires that travel time increases must adversely impact productive activity before it is appropriate to assign monetary values to them. As it is implausible that the small daily increases in travel time resulting from lower speeds on urban local streets have any measurable impact on productive activity, and as it is unlikely that any individuals will ever be faced with long delays as a result of the lower speeds, calculation of monetary costs of increased travel time would be inappropriate.” (Austroads, 1996, p.21) Furthermore, because the average increase in travel time is of the order of 4-10%, such impacts fall within the normal range of variability of urban trips and, therefore, are unlikely to be noticed by vehicle occupants. The NSW preliminary evaluation found that 25% of persons interviewed did not perceive an increase in travel time and 41% considered it to be slight (ARRB Transport Research 1999, cited in VicRoads, 2000). Given this, vehicle occupants are unlikely to place a high value on travel time increases of this order. Page 60 Evaluation of a 50 km/h Default Urban Speed Limit for Australia 5.5.2 Effects of methods of implementation The outcomes summarised in Tables 35 and 36 are not predicated on whether the 50 km/h urban speed limit is implemented by default or by signing of speed limited areas. There is currently no clear evidence regarding the relative effectiveness of these two approaches to implementation. The summaries do not incorporate implementation costs which are likely to vary significantly according to the method and extent of implementation. As noted earlier in the report, most States and Territories will have implemented a 50 km/h urban speed limit for local streets in some form or another by the end of 2001. Therefore, any estimate of implementation costs based on an assumption of a 60 km/h national baseline for local streets is largely hypothetical. However, it can be assumed that there has been no widespread implementation of 50 km/h speed limits on urban collector roads and arterial roads. The real cost of implementation of a default 50 km/h urban speed limit in local streets would depend on whether those States and Territories that have already adopted an area-wide approach would change to the default approach. There is no strong evidence of any additional road safety benefits of a uniform approach to implementation of 50 km/h urban speed limits (although as discussed in Section 8 attention to key aspects of planning, coordination and implementation can contribute to how effectively the change to a lower limit is introduced). The set of options for implementation includes: Option 1: the hypothetical situation of implementation of a default limit assuming 60 km/h on local streets throughout Australia Option 2: implementation of a default limit from current situation (assuming change to a default limit) Option 3: implementation of a default limit from current situation (States and Territories with signing of speed limited areas remain the same) Option 4: the hypothetical situation of implementation of signing of speed limited areas assuming 60 km/h on local streets throughout Australia Option 5: implementation of signing of speed limited areas from current situation (assuming change to signing of speed limited areas) Option 6: implementation of signing of speed limited areas from current situation (States and Territories with a default limit remain the same) The costs of implementation of lower urban speed limits on local streets by changing default speed limits have been estimated as $2.8 million in Victoria and about $2 million in Western Australia. VicRoads (2000) estimated that the costs of extensive signing to achieve the equivalent outcome would have been almost five times higher. On the basis of these estimates, the hypothetical implementation assuming a 60 km/h baseline across Australia would cost in the order of $30 million for a default limit (Option 1) and in the order of $150 million for implementation by signing (Option 4). If States and Territories that had already adopted 50 km/h residential speed limits did not change their method of implementation, then the total implementation costs would be less than the estimates for the hypothetical case. Thus Option 3 (implementation by default limit) would cost less than about $30 million and Option 6 (implementation by signing of speed limited areas) would cost less than about $150 million. Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 61 The implementation costs would be greater if States and Territories that had already adopted 50 km/h residential speed limits changed their method of implementation. It is more difficult to estimate the relative cost of the national uniformity options (Options 2 and 5). Page 62 Evaluation of a 50 km/h Default Urban Speed Limit for Australia 6. CONSULTATION 6.1 Consultations undertaken Relevant agencies in each State and Territory were advised of the purpose and scope of the project and assistance was sought for information to help meet its aims. However, broad-based consultations with a wide range of stakeholders did not form part of the evaluation phase of the project. Such consultations will be initiated during the second half of 2001. They are expected to provide valuable feedback on the form of assessment undertaken and on practical issues associated with any future implementation of 50 km/h limits. As part of such consultations, consideration should be given to a meeting between State and Territory representatives - in a workshop format - at which the merits of different ways of implementing a 50 km/h limit could be explored in detail. Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 63 7. REASONS FOR PREFERRING PROPOSED APPROACH 7.1 Decision criteria for a 50 km/h speed limit The focus of the analysis in this report goes beyond the justification of a regulation per se (as a default limit of 60 km/h in built-up areas already exists in the Australian Road Rules) with the assessment, on the basis of available evidence, of whether a change in the current limit is desirable. The primary consideration in any decision to reduce speed limits is whether a lower limit reduces speeds and to what degree. A range of international studies suggests that reductions in speed are modest but real (Leaf and Preusser, 1999). More importantly, even small reductions in prevailing vehicle speeds result in reduced crashes and injuries. In the urban context, this is particularly significant for vulnerable road users, especially pedestrians. The initial results of the introduction of 50 km/h limits in New South Wales and Queensland reviewed in this report support the conclusion that a reduction in the limit lowers speeds and crashes. Associated considerations include the positive and negative impacts of lower speed limits on the community more broadly, in areas such as travel times, vehicle operating costs, vehicle emissions and the amenity of urban environments. This report finds that the major benefit would be fewer casualty crashes. Fewer property damage only crashes and reduced air pollution would be minor benefits. The major cost would be increased travel time and a minor cost would be increased vehicle operating costs. The net outcome depends on how meaningful it is to value very small increases in travel times. If these are valued, then a reduction in the default urban speed limit to 50 km/h is economically justified only for urban arterial roads currently zoned 60 km/h. If the small travel time increases are not valued, then a reduction in the default urban speed limit is economically justified for all classes of road considered (local streets, collector roads and urban arterial roads currently zoned 60 km/h). The savings in casualty crash costs exceeded the savings in property damage only crash costs and modest benefits were identified from reductions in vehicle emissions. There were several factors that led the estimation of benefits to be conservative. First, the speed-related impacts of carbon dioxide and noise emissions were not measured. However, since these increase with speed, the impact of a lower speed limit in this area would be positive. Secondly, the possible benefit of improved speed compliance on collector and arterial roads resulting from lower limits on local streets was not able to be measured. Thirdly, the estimates of crash numbers were based on reported crashes only and therefore the benefits in reductions of non-reported crashes are not included. The analysis is probably made more conservative by over-estimation of costs. The travel time increases are likely to be overestimated because they do not take into account route substitution, destination substitution, or trip suppression effects. With regard to travel time impacts, the estimated average increase per head of population in Australia ranged from about nine seconds per trip up to approximately 25 seconds per trip. If Australians were to accept travel time impacts of this order, it is estimated that between 2,900 and 7,380 casualty crashes would be prevented in Australia each year. Environmental and health considerations have provided impetus for the encouragement of walking and cycling as forms of transport. Given the key influence of speed on the severity of injuries sustained by pedestrians and cyclists, this has the potential to expose a greater number of vulnerable road users to risk. Managing the speed of vehicles by appropriate speed Page 64 Evaluation of a 50 km/h Default Urban Speed Limit for Australia limits goes hand in hand with the higher priority being given to non-motorised forms of travel. Other approaches can be seen as both alternative and complementary ways to reduce speeds in urban areas. The effectiveness of these alternatives varies from low (for example public education) to significant (for example traffic calming) with the latter being a higher cost alternative, and requiring application over the longer term to impact on a significant proportion of the urban environment. A characteristic of the regulatory approach exemplified by a default limit is its capacity to have an immediate impact across a whole population, the effectiveness of which can be periodically reinforced by associated measures such as enforcement in combination with targeted public education programs. 7.2 Recommendation It is recommended that national consideration be given to the adoption of a 50 km/h default urban speed limit in the Australian Road Rules. 7.3 National competition policy The Competition Principles Agreement sets out the basic principle that must be applied to legislation, namely, that it should not restrict competition unless it can be demonstrated that: • the benefits of the restriction to the community as a whole outweigh the costs • the objectives of the regulation can only be achieved by restricting competition. A reduction in the urban default speed limit does not directly provide for, nor have the effect of, a restriction on competition. There are no features of such a decision that impose barriers to entry or restrictions on competitive conduct. The direct effect is on how roads are used, not who can use them (VicRoads, 2000). Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 65 8. IMPLEMENTATION 8.1 Issues affecting the implementation of 50 km/h limits The analysis in Section 5 of the benefits and costs of a 50 km/h default urban speed limit showed that the implementation is economically justified only for urban arterial roads currently zoned 60 km/h. If small increases in travel time are not valued, then a reduction in the default urban speed limit is economically justified for all classes of road considered (local streets, collector roads and urban arterial roads currently zoned 60 km/h). On the other hand, jurisdictions have so far chosen to apply the lower limit essentially to local streets. Depending on the jurisdiction, this has been achieved by a default limit for the affected streets (Statewide or on a zonal basis) and the use of signing either to delineate the affected areas or to indicate where a higher speed limit applies (such as on collector roads). The lessons to be drawn from Australian experience in regard to the implementation of 50km/h limits therefore focus on the more limited context of local streets. Should a jurisdiction undertake implementation of a 50 km/h limit on a broader scale, encompassing segments of urban arterial roads, the challenge would be greater and different approaches may be required. Experience to date suggests that the steps taken to plan, coordinate and implement a 50 km/h speed limit are likely to influence its effectiveness in reducing speeds and crashes. Key aspects of implementation discussed in this section include: • the roles played by State and local governments; • identification of affected roads; • informing the public of change; • achievement of compliance, and promotional support for enforcement. 8.1.1 Respective responsibilities of State and local governments for implementation Generally, State and Territory Government agencies are fully responsible for the funding and management of State Highways, and are either fully responsible or share, with local government, funding and management responsibility for main roads (generally corresponding to the arterial category). Local governments are typically fully responsible for the funding and management of local roads (Austroads, 2000). Experience indicates that planning, coordination and integration are essential features of effective implementation of new speed limits. Donaghey and Ram (2000) observed that: “The success of the 50 km/h local street speed limit initiative in south-east Queensland has been due largely to the integrated implementation program that has been adopted. It was recognised that an effective implementation would require much more than simply erecting a few signs. The implementation program would need to be based on the ‘Three Es’ - Education, Engineering and Enforcement. The implementation program required a working partnership between Queensland Transport, the Queensland Police Service, each of the local governments in south-east Queensland, and the Department of Main Roads.” Page 66 Evaluation of a 50 km/h Default Urban Speed Limit for Australia In the case of New South Wales, Rouse (2000) observed that: “The key factor contributing to the success of implementing the 50 km/h limit was the partnership between councils, their communities, the police and the RTA. This component is followed by clear, succinct technical guidelines, a well planned and executed public education campaign and an extensive evaluation of the lowered limit.” The principle of co-ordinated and integrated action had previously been re-affirmed by various inquiries/task forces that have addressed the introduction of lower limits in local streets: • The New South Wales Parliamentary Joint Standing Committee on Road Safety (Staysafe) in the report of its inquiry (1996) recommended that the speed limit change as one action within an integrated package of measures which would include traffic management, traffic law, police enforcement, and publicity strategies; • The Austroads report into Urban Speed Management in Australia (1996) concluded that a number of steps were needed to achieve lower speeds in local streets, including: - all required speed signs should be installed prior to any regulation changes; - particular attention should be paid to publicity, education and enforcement strategies in implementing change; - community and interest group input on the change should be encouraged via public discussion papers; - adequate monitoring and evaluation of the impacts of changes to urban speed management on vehicle speeds, accidents, travel times and amenity should be carried out. The direct costs of implementing a 50 km/h limit are associated with the required signing, public education, enforcement of compliance, and monitoring and evaluation tasks. The arrangements for sharing these costs vary between jurisdictions but typically the greatest proportion has been met by the relevant central agencies. Conclusion The effective implementation of 50 km/h speed limits is facilitated by establishing procedures that ensure that an integrated and coordinated approach between responsible State, Territory and local governments is undertaken from an early stage. 8.1.2 Identification of affected roads Central to the effective implementation of a lower limit in local streets is the identification of the streets to which the lower limit will apply. Jurisdictions have approached this in a number of ways. In New South Wales the Roads and Traffic Authority provided councils with guidelines for implementing the 50 km/h speed limit that outlined key steps and issues in the development of road hierarchy plans and the installation of signs. The initial guidelines were later revised in Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 67 consultation with councils. These plans were developed jointly by the authority and councils to ensure that streets are zoned consistently and minimise the number of speed zone changes. In Queensland, the 50 km/h speed limit was applied to local streets in the south-eastern area in accordance with the Queensland Manual of Uniform Traffic Control Devices to prevent inconsistent speed zoning across local government areas. This involved working closely with local government to determine the application of the limit. The Manual listed typical characteristics of local streets for guidance. Consistency and credibility were re-enforced by requesting local government to submit speed zone plans before implementation. A 50 km/h limit was also applied in some situations other than on local streets. Examples include strip shopping centres, foreshores and where the physical environment supports the lowered limit. In these particular cases the 50 km/h limit is signposted. In Victoria any street in a built-up area without a sign automatically has a 50 km/h limit. The issue in Victoria was the identification of those local streets where a 60 km/h limit was to be retained and signs needed to be erected. In conjunction with the adoption of the 50 km/h default limit, VicRoads was given authority to install speed signs on local roads in addition to its existing powers on declared roads. The main purpose of this additional power was to help ensure that collector roads are sign posted in a consistent manner across local government areas. The erection of these signs was arranged by VicRoads in consultation with local government. In the Australian Capital Territory. the 50 km/h speed limit applies to suburban streets that are used to provide access to homes and carry only neighbourhood traffic. Major roads that pass through suburbs continue to have a 60 km/h limit. Roads in commercial centres, industrial areas and the Parliamentary zone also retain the previous limit. Western Australia will implement the 50 km/h limit for local streets across the State, using criteria developed by Main Roads WA. The aim is to apply the limit on local roads that carry neighbourhood traffic (local distributors) or give direct access to properties. In Perth, where there is an agreed functional road hierarchy, all local roads not signed at 60 km/h (or other limits) will be 50 km/h by default. In rural areas and major regional centres, similar principles to those intended for Perth will be used to determine the local roads to which the limit will apply. Conclusions • The selection of roads to which a 50 km/h limit should apply, along with the signing of those roads that need to retain a 60 km/h limit, should be undertaken jointly by the central road agency and local government; • Sufficient time should be allowed for the effective completion of this process to enable a smooth transition to the new speed limit structure; • Effective implementation processes help to achieve public acceptance of change and the retention of community support for speed management initiatives undertaken in the future. Page 68 Evaluation of a 50 km/h Default Urban Speed Limit for Australia 8.1.3 Informing the public of change A change in the speed limit is a major one and not frequently undertaken. If insufficient effort is made to communicate the purpose and nature of the change, it represents an opportunity foregone to influence public perceptions and road user support. All jurisdictions that have implemented the 50 km/h limit have identified community education as a key factor and undertaken activities relevant to their needs. For example: • New South Wales: A public information campaign was developed with the following elements: - participating councils could apply for grants for community education and consultation programs so that more specific localised strategies could be conducted to enhance the Roads and Traffic Authority program; - conduct of a “generic” campaign in all councils implementing the new limit incorporating newspaper advertisements and a brochure to all households; - television and radio advertising campaigns targeted to specific regions of the State and also more broadly across the State; - a range of other initiatives such as an information hotline, posters, a media information kit, and material provided to key stakeholders such as the NRMA for use in its publications with wide circulation. • Queensland: A public education campaign was needed to target south-east Queensland residents as well as visitors travelling to the region from other parts of the State, and interstate and overseas visitors; the campaign included: - a mass media campaign; - a brochure and map mailed directly to households in south-east Queensland; - promotions and educational displays; - marketing within local government areas by council officers; - marketing activities targeting visitors to the area. Aspects of communication were reviewed on the basis of community feedback and amended to ensure that any confusion about the application and impact of the lower limit was overcome. • Victoria: An educational campaign was undertaken by VicRoads at the time of introduction of the default limit under the umbrella message of “Think safe. Think 50”. It concentrated on radio and print media. Brochures were also distributed to households. There was extensive editorial coverage in the print and electronic media. The Transport Accident Commission provided support by mounting a campaign focused on television. Queensland experience shows the importance of the careful choice of actual content of public communication in order to gain community support (Donaghey, Ram, 2000). The public was told that the speed limit would reduce to 50 km/h on around 90% of the streets in built-up areas. This had the opposite effect of that intended with many people assuming that they would be required to drive at 50 km/h for the majority of the time. Communication was Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 69 adjusted to emphasise the fact that drivers would generally be travelling at 50 km/h for only one or two minutes at the start and end of each trip. Conclusion A range of media and public education activities should be planned and conducted as appropriate on a statewide and local basis, in specific regions and directed to relevant groups in the community to ensure community awareness of the intended change in speed limits. 8.1.4 Achievement of compliance Speed limits, along with many other aspects of road law, suffer from the disadvantage of not being self-enforcing. On the other hand, public support for a 50 km/h speed limit in local streets is high, along with support for enforcement of limits seen to be reasonable and appropriate. All jurisdictions that have implemented 50 km/h limits have accepted that enforcement is a necessary part of making lower limits effective. The attitude of responsible agencies has been that police will enforce the limit as part of their overall traffic safety role. Generally, enforcement resources will be utilised in accordance with each jurisdiction’s policing strategies, for example, using crash data and information systems to guide enforcement tactics. The position of jurisdictions can be summarised as follows - “People who drive above the speed limit are likely to receive an infringement notice - just as they would if they drove above 60 km/h now” (Kidd, 2000). Conclusion Enforcement is a necessary part of ensuring maximum compliance with a lower speed limit and should be undertaken with sufficient intensity to achieve the desired change in road user behaviour. 8.1.5 Promotional support for enforcement In addition to making the community aware of the reasons for change in the speed limit and its specific application through a program of public education, promotional support for enforcement using appropriate mass media should be seen as an integral part of achieving compliance. This is strongly supported by evidence that the combination of legislative change, rigorous enforcement backed by promotional support can be instrumental in changing road user behaviour (Cameron, Haworth, Oxley, Newstead, Le, 1993). Conclusion Enforcement of a change in the speed limit should be supported by promotion using appropriate mass media to maximise its impact on road user behaviour. Page 70 Evaluation of a 50 km/h Default Urban Speed Limit for Australia 8.2 Guidelines for implementation It is not possible to conclude on the basis of existing information whether there is a specific preferred implementation model. There are both similarities and differences in the approaches taken to date. An evaluation of the Victorian approach - which unlike some other jurisdictions relies predominantly on regulation without major expenditure on signing - is not yet available. This precludes an objective comparison. Nevertheless, the experience so far points to a number of important, and in some cases fundamental, steps that should accompany any future decision by a jurisdiction to adopt a 50 km/h limit. The successful implementation of a 50 km/h speed limit regime is reinforced by managing the process in a co-ordinated and integrated manner, with an emphasis on: • ensuring that appropriate planning takes place among the responsible central agencies and local government from an early stage; • collaboration between the central road agency and local government in the selection of roads to which a 50 km/h limit should apply, along with the signing of those roads that need to retain a 60 km/h limit; sufficient time should be allowed for the effective completion of this process to enable a smooth transition to the new speed limit structure; • giving priority to a structured and managed approach to help achieve public acceptance of change and the retention of community support for any future speed management initiatives; • planning and conducting a range of media and public education activities on a statewide and local basis, in specific regions and directed to relevant groups in the community to ensure community awareness of the intended change in speed limits; • conducting enforcement as a necessary part of ensuring compliance with a lower speed limit, undertaken with sufficient intensity to achieve the desired change in road user behaviour; • providing promotional support for enforcement using appropriate mass media to maximise its impact on road user behaviour. Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 71 9. REFERENCES Australian Bureau of Statistics. (2000). Survey of Motor Vehicle Use. 12 months ended 31 July 1999 (Cat No. 9208.0). Canberra: ABS. Austroads (1995) Guide to traffic engineering practice: Part 13 - pedestrians. Sydney: Austroads. Austroads (1996). Urban speed management in Australia. Report AP 118. Sydney: Austroads. Austroads (2000). RoadFacts 2000. Sydney: Austroads. Brindle, R. (1998) Managing the impacts of traffic on local communities. Proceedings of the National Road Safety Summit, 16-18 September 1998, Canberra: Federal Office of Road Safety. BTE. (2000). Road crash costs in Australia (Report 102). Canberra: Bureau of Transport Economics. Cameron, M. (2000). Estimation of the optimum speed on residential streets. Draft report to Australian Transport Safety Bureau. Cameron, M., Haworth, N., Oxley, J., Newstead, S. and Le, T. (1993). Evaluation of Transport Accident Commission road safety television advertising. Melbourne: Monash University Accident Research Centre. Corben, B. and Diamantopoulou, K. (1996) Pedestrian safety issues for Victoria. Melbourne: Monash University Accident Research Centre. Dyson, C.B., Taylor, M.A.P., Woolley, J.E. and Zito, R. (2001). Lower urban speed limits - trading off safety, mobility and environmental impact. Paper presented at the 24th Australasian Transport Research Forum, 17-20 April, Hobart. Donaghey, B.and Ram, S. (2000). Effective speed management. 50 km/h local street speed limit – An engineer’s perspective. Paper presented at the Public Forum on 50 k/h Speed Limit for Local Roads. http://www.transport.wa.gov/roadsafety/issues/donaghey.htm. Engel, U. and Thomsen, L. (1992). Safety effects of speed reducing measures in Danish residential areas. Accident Analysis and Prevention, 24, 17-28. Haworth, N. and Symmons, M. (2001). Safety and environment in Vision Zero. Paper presented at the 24th Australasian Transport Research Forum, 17-20 April, Hobart. Haworth, N., Tingvall, C., Vulcan, P. and Cameron, M. (1999). Future directions: “Safety First” 2000-2005. Development of road safety initiatives. Report prepared for VicRoads. Jorgensen, E. (1994). The effects of changed urban speed limits. Permanent International Association of Road Congresses Technical Committee C13 – Road Safety. Kallberg, V-P. and Toivanen, S. (1998). MAnaging Speeds of Traffic on European Roads. Application of the MASTER framework. Link-level analysis of the impacts of a speed management policy. http://www.vtt.fi/rte/projects/yki6/master/blankins.htm Page 72 Evaluation of a 50 km/h Default Urban Speed Limit for Australia Kidd, B. (2000). 50 km/h urban speed limits: The Western Australian approach. Paper presented at the Public Forum on 50 k/h Speed Limit for Local Roads. http://www.transport.wa.gov/roadsafety/issues/norwell.htm. Kloeden, C.N., McLean, A.J., Moore, V.N., and Ponte, G. (1997). Travelling speed and the risk of crash involvement (CR 172). Canberra: Federal Office of Road Safety. Leaf, W.A. and Preusser D. F. (1999). Literature review on vehicle travel speeds and pedestrian injuries. US Department of Transportation, National Highway Traffic Safety Administration, DOT HS 809 021. Market Equity Pty Ltd. (1994). Leanyer Speed Trials Community Survey. Report to Darwin City Council. Fannie Bay, NT: Market Equity Pty Ltd. McLean, A.J., Anderson, R.W.G., Farmer, M.J.B., Lee, B.H. and Brooks, C.G. (1994) Vehicle travel speeds and the incidence of fatal pedestrian collisions. Volume 1. (CR 146). Canberra: Federal Office of Road Safety. Meers, G. and Roth, M. (2001). Road safety and ecological sustainability working together. Paper presented at the 24th Annual Conference of the Transport Research Forum, Hobart. Mitchell-Taverner, P. (2000) Community attitudes to road safety: Community Attitudes Survey Wave 13. Canberra, Australian Transport Safety Bureau. Newstead, S. V., Cameron, M. H. and Le, C.M. (2000). Vehicle crashworthiness and aggressivity ratings and crashworthiness by year of vehicle manufacture: Victoria and NSW crashes during 1987-98, Queensland crashes during 1991-98. Melbourne: Monash University Accident Research Centre. NFO Donovan Research (2000a). Attitudes to the proposed 50 km/h speed limits on local area roads. Report to Office of Road Safety, WA. NFO Donovan Research (2000b). Community Attitude Monitor. Report to Office of Road Safety, WA. NFO Donovan Research (2000c). Community Attitudes Survey – Speed limits on local area roads. Report to Office of Road Safety, WA. Page, Y. (1993) The implementation of 50 km/h in towns and its effects on road safety. Transport Safety Research No 41, December 1993. Preston, B. (1990). The safety of walking and cycling in different countries. In Tolley, R. (ed.), The greening of urban transport: Planning for cycling and walking in Western cities. London: Bellhaven Press. Radalj, T. (1999). Cited in - 50 km/h speed limit on local roads: questions and answers. Office of Road Safety 50 km/h webpage. http://www.transport.wa.gov.au/roadsafety. Road Safety Council of Western Australia. Cited in - Speed: the facts. Office of Road Safety 50 km/h webpage. http://www.transport.wa.gov.au/roadsafety. Road Safety Council of Western Australia. (2001). Road Safety Strategy. Rouse, R. (2000). Implementation of the 50 km/h urban speed limit in NSW. Paper presented to the Public Forum on 50 km/h Speed Limit for Local Roads. Evaluation of a 50 km/h Default Urban Speed Limit for Australia Page 73 RTA. (2000a). 50 km/h urban speed limit evaluation. Summary report. Sydney: Roads and Traffic Authority. RTA (2000b). Sharing the main street. Sydney: Roads and Traffic Authority. RTA (2000c). Road traffic accidents in NSW – 1999. Sydney: Roads and Traffic Authority. SMEC Australia. (1998). Effects of urban speed management on travel time. Working Paper 2: Micro trip analysis. Report to Federal Office of Road Safety. STAYSAFE 34 Parliament of New South Wales Joint Standing Committee on Road Safety (1996). A 50 km/h general urban speed limit for New South Wales. Report 7/51. Taverner Research Pty Ltd. (2001). ACT speed limit trial survey. Report to ACT Dept of Urban Services. Thoresen, T. (2000). Economic evaluation of road investment proposals: Unit values for road user costs at June 1997 and June 1998 (AP-142/99). Sydney: Austroads. VicRoads (2000) Regulatory impact statement for proposed Road Safety (Road Rules) (Amendment) Regulations 2000. Melbourne: VicRoads. Walsh, D. (1999). The 50 km/h speed limit in New South Wales: A joint partnership between councils, their communities and the NSW Roads and Traffic Authority. Paper presented to 1999 Research, Policing, Education Road Safety Conference. (pp.695-707) Walsh, D. and Smith, M. (1999). Effective speed management The next step forward: Saving lives by decreasing speeds in local streets. Paper presented to 1999 Research, Policing, Education Road Safety Conference. (pp.685-694) Walz, F.H., Hoeflinger, M. and Fehlmann, W. (1983). Speed limit reduction from 60 to 50 km/h and pedestrian injuries. In Twenty-seventh Staff Car Accident Conference Proceedings with International Research Committee on Biokinetics of Impacts (IRCOBI), pp. 311-318. October 1983, San Diego, CA Warrendale (PA): Society of Automobile Engineers. Ward, H., Robertson, S. and Allsop, R. (1998). Managing speeds of traffic on European roads: Non-accident external and internal effects of vehicle use and how these depend on speed. Paper presented at Workshop II on Speed Management, Proceedings, 9th International Conference on Road Safety in Europe, Cologne, Germany. Western Australian Municipal Association (2000). Proposed 50 km/h urban speed limit – Local government survey. Perth: WAMA." />
  • Media Releases
  • Media Clippings
  • Advertisement
  • Issues and Policies
  • Reports
  • Links

Prince Charles - (QUOTE):
"The whole of the 20th century has always put the car at the centre. So by putting the pedestrian first, you create these liveable places I think, with more attraction and interest and character ... liveability."
click here to view
Follow us on Facebook

Evaluation of a 50 km/h Default Urban Speed Limit

National Road Transport Commission

November 2001
Report Prepared by: Monash University Accident Research Centre

Authors: Narelle Haworth, Bob Ungers (Bob Ungers Consulting), Peter Vulcan, Bruce Corben

Click here to download document (pdf format)