LBC SCRUTINY MEETING - 20MPH LIMITS - PJM REPORT – 14-9-16
I am a local resident with extensive knowledge of the road network of Croydon.
For well over 20 year I have actively monitored and analysed council traffic and parking schemes, read TMCC reports and appeared at committee.
I was a member of council committees for traffic management and road safety from 1994 until 2008, and NP rep for roads and traffic for Coulsdon from 2005-2009.
These positions ceased when the previous council withdrew support for them.
There is nothing intrinsically good in reducing vehicle speeds.
This may be necessary on safety grounds, but slower journeys inevitably impose adverse economic effects.
The council report states that accidents and casualties are currently around the lowest ever, and are much lower than 10 and 20 years ago. This of course is to be welcomed.
What is absent is any real knowledge or analysis of why this has happened.
A colleague of mine has produced the report which I have submitted. Both he and I have degrees in Mathematics including data analysis.
By profession, he is a senior bus scheduler for a major London bus company, operating a large number of bus routes with hundreds of employees. He is responsible for analysing bus journey times, and drawing up new schedules to reflect traffic conditions amongst other factors.
Thus when he writes “I see the enormous damage done to bus service provision and reliability by the introduction of traffic schemes. Bus journey times in London continue to rise by 1-2% per year, despite overall traffic levels falling, and I believe this is almost entirely due to the way the road network is being managed. There is much talk about a 20mph London-wide limit, and this alarms me enormously - bus services may have to be cut by 20-30%, unless a massive increase in subsidy is forthcoming.”, he is speaking with expert knowledge of the actual impact on daily bus journeys of road network management.
Clearly the impact of 20mph limits on bus services will depend on which roads are affected.
The council report says that major A roads will not be included, but that B roads and lesser roads could be subject to 20 limits.
Many bus routes in Croydon use these roads, and would be impacted.
Of course this applies to general traffic as well, with some journeys of several miles on such roads
In terms of the actual effect on road safety, the report I have submitted analyses the improvement in accidents and injuries across London boroughs over the past 20 years.
The trends have been broken down according to the extent that 20mph zones and other traffic calming measures have been implemented.
The study shows clearly that boroughs with a high level of 20 zones have achieved a LOWER reduction in casualties than those with a low level. The amount of data here is statistically significant.
I would add to this that boroughs with a much increased level of cycling have seen a significant increase in injuries and deaths in that category.
This should be compared with the fact that nationally since 1990 there has been a 20% RISE in all traffic, yet a 60% REDUCTION in KSI and a 35% reduction in slight injuries.
The report refers to the claim that “research has shown that a pedestrian has a 90% chance of surviving an accident at 20mph, as against 50% at 30mph”.
I challenge any such assertion.
I have researched the basis for these claims via FOI to the DfT, and they appear to be based on one analysis from many years ago, which attempted to back calculate from estimated speeds in fatal accidents.
Noone has referred to any scientific study which has tested, with dummies, outcomes after being struck at various speeds.
Any calculations made on fatal accidents are necessarily unreliable, as they rely on estimates for vehicle speeds which are typically say between 35 and 43 mph. They are also unrepresentative, as most injury accidents are not fatal.
Accidents which result in serious injury are analysed less fully by the police, and minor injury accidents may not really be analysed at all.
The police are also primarily interested in whether an offence has been committed, rather than a full assessment of the causation of an accident.
Thus there is in fact no reliable data for outcome in terms of injury or death for pedestrians struck by vehicles at varying speeds, and so ANY CLAIMS ARE UNRELIABLE and MUST NOT BE RELIED UPON.
Furthermore, we know there have been many cases of someone being knocked over backwards, for example by a punch, and being killed when their head strikes the ground. That this is not due to the force of the punch as such, but to the fall, may readily be seen by the way some can take a punch without collapsing.
The outcome when a pedestrian is struck by a vehicle depends on the point of impact. A pedestrian struck by a large vehicle with a flat front, eg a bus or lorry, even at slow speed, may be knocked over backwards and die. Their head may also be directly impacted. An adult pedestrian hit by a car will typically fall forward onto the bonnet, as they are struck in the legs. Injury results more when they land on the ground. The outcome depends critically on how they strike the ground rather than the speed they are moving at. For example, a motor cyclist with suitable clothing may crash at 60mph, slide along the ground, and get up uninjured. However if he crashes at 20 and slides into a tree, he may be killed.
Certainly I accept that as a general rule, being struck at a higher speed may on average be expected to lead to more serious injury.
However we have no credible basis for assessing the extent of any such tendency.
The Portsmouth study claims an accident reduction of just under 19%, looking at 3 year periods before and after it put in its 20mph zones widely across the borough. It refers to all injuries, claiming the KSI data is too small to be reliable,
The report I submitted quotes KSI data, as that is generally regarded as the most relevant.
This shows that London achieved a reduction of 21% in KSI over the same periods as Portsmouth.
The Portsmouth effect may be compared with the national outcome reduction of around 20%.
Thus Portsmouth achieved no better a reduction in slight injuries than was observed across the country. We may note also they do not quote control periods before and after, eg 01-03 for 05-07 and 12-13 for 09-11.
There is thus no reason to regard the 20mph scheme in Portsmouth as particularly successful.
The council report is weak on the negative sides of 20mph limits.
The report asserts that “The principal aims of any 20mph speed limit are to improve road safety, discourage through traffic, encourage walking and cycling, make a positive contribution to improving health and tackling obesity and improve the quality of life and the local environment.”
We can measure changes in casualties from accident data. However causality is not thereby established.
Notably there is also police STATS 19 data, which includes assessment of causes.
In terms of “discourage through traffic”, it is unclear why this is an objective, or how any success it might achieve might be assessed. For example, along the A23 corridor between Purley and Coulsdon, use of side roads by motor traffic is essential to the effective functioning of the main road. If traffic were persuaded by 20 limits to switch to the A23, there would be severe adverse impacts on that road, and notably on buses.
As a result, any switching is likely to be minimal.
It is totally unclear how a 20 limit might “encourage walking and cycling”, nor how any success in that might be measured.
Accident data shows clearly that increased cycle use, unlike increased car use, leads to increased casualties. Thus any council that actively promotes cycling is knowingly actually increasing the risk profile of the road network.
I do not say that people should not cycle. If they wish to take the risk, that is their right, and the road network must be made as safe as possible for all road users.
However for any council to actively encourage cycling is surely irresponsible.
To suggest that a 20 limit might “make a positive contribution to improving health and tackling obesity” is surely fanciful – apart from any effect on road safety.
Many people drive to the gym, so while brisk walking for some considerable distance can be beneficial, so can going by car to a workout.
Turning to “improve the quality of life and the local environment”, we should look at the negative aspects to a 20 limit – and it is hard to see how a 20 limit could achieve a net positive outcome.
Travelling a mile at 30 take 2 minutes. At 20 it takes 3 minutes, a 50% increase in journey times.
We should note that the 10mph reduction is far greater than the effect of going at 70 rather than 80mph – here the effect is only around 13%. Yet the government looked at raising the motorway speed limit as "Increasing the motorway speed limit to 80mph would generate economic benefits of hundreds of millions of pounds through shorter journey times” http://www.bbc.co.uk/news/uk-politics-15116064
It is surely clear that a 13% journey time reduction would have these benefits, then a 50% increase would have much greater adverse effects.
We must not fall into the trap of arguing that because residential side streets are short and traffic levels are low, the effects of 20 rather than 30 are negligible.
Many journeys on side roads are several miles long. Going by main road would often be much further in distance and time.
The effect of widespread 20 limits implies a major increase in journey times. It may be only a minute on one journey, but when that is added up over thousands of journeys every day, the overall economic impact is significant.
Yet the council fails to quantify at all this effect.
Consider a one mile length of road, with a traffic flow of 6 vehicles a minute, typical for many unclassified link routes in the rush hours. At 30 it takes 2 minutes and at 20 it takes 3 minutes. Thus at 30 you will have 12 vehicles in each direction on the road. Slow it down to 20, and you now have 18 vehicles each way.
It is not just an increase of 50% in journey times that results, it is also an increase of 50% in traffic on the stretch of road
For a pedestrian wishing to cross the road, there will be a car coming close by more often. Yes it will be going more slowly, but the pedestrian still has the problem of whether to cross with a car approaching.
As a clear rule, the faster a car is going the more stable the speed is – as opposed to the vehicle itself.
The faster a car is going, the more likely it is to continue in a straight course and at a similar speed.
At 20 rather than 30, a car may vary its speed and direction much more suddenly.
This is a real problem for a pedestrian wishing to cross the road.
For example, going at 30, a car needs to slow significantly to turn left or right.
At 20, it merely needs to ease off to say 15 to make the turn.
Thus slowing vehicles makes it harder for a pedestrian to predict what they will do, so more than cancelling out any benefit from the car going more slowly (in the context of 20 and 30).
Thus traffic at 20 rather than at 30 can make it actually harder to cross the road safely.
As a pedestrian, I prefer traffic to go at 30 rather than 20, other things being equal.
A car is at its mist efficient in terms of fuel economy around 35mph, and this falls off as the speed rises over 40 and as it drops under 30,
At 30, a car will be gong in 4th or 5th gear.
At 20, it will probably be 3rd gear.
Cars going at 20 rather than 30 will use much more fuel and put out more CO2 and pollutants per mile.
The faster a vehicle moves the more wind displacement results.
Cars going at 20 rather than 30 means more concentrated pollutants, as there is less wind dispersion and dissipation generated.
Consider a house with a 30m frontage. At 30, a car passes in a fraction over 2 seconds. At 20, it takes 3 seconds. Thus the house is exposed to direct pollution from the car for 50% longer. It will also take longer to dissipate away from the house.
In terms of noise, going at 20 rather than 30 may mean less, the same, or more noise, depending on a range of factors. Engine noise is a major factor, notably for buses and lorries.
In 3rd gear at 20 rather than 4th gear at 30 means similar engine revs and similar noise emitted.
So the house will get a similar level of traffic noise, but for 50% longer for each vehicle going at 20 rather than 30.
Portsmouth claim that traffic speeds before their 20mph innovation were typically around 20-25mph, and this then dropped overall by a fraction over 1mph when they reduced the limit to 20.
They claim a range of reductions up to 6mph. However DfT guidance states that where there are no other speed reducing features, only a 1mph reduction may be expected.
This is of course an average.
If 1 car in 10 has a seriously law-abiding driver, they will cut their speed from 27-30 to 17-20.
If the overall effect is a 1mph average reduction, then that means the other 90% continue at the same speed as before.
Scientific analysis of safety and speed data has shown that it is safest if everyone drives at about the same speed. Thus if everyone is doing around 35 in a 30 limit, that is safer than 90% doing around 35 and 10% doing 30 or below.
Thus the 20 limit may actually decrease safety by increasing the range of speeds of traffic.
On the Portsmouth data, before the 20 limit, the vast majority of drivers were complying with the 30 limit.