Manchester Found Casualties Falling in its 20mph Areas

Manchester reported on their first 20mph limits. It resolved to complete Phase 2 of the 20 mph programme, continue supporting phases 1 and 2 of the 20mph scheme and monitor and review data to increase its understanding. It incorrectly compared % falls for small number samples in already safer than average areas versus citywide.  

Reporting to Manchester’s Executive on 8th March 2017, officers gave figures on road casualties and speeds. See (item 16)

Casualty effects comparing 2012/14 with 2014/16 were presented for Phase 1 - Gorton North & South, Miles Platting & Newton Heath, Moss Side & Fallowfield.  It says “Overall the results show that casualties in the phase 1 20mph area have not reduced as much as the casualty numbers citywide. We would need to evaluate data over a longer time to get data that is more statistically relevant.”  20’s Plenty for Us agree.  Absolute numbers are small, for instance total pedestrian casualties in Phase 1 fell from 53 to 42 (11 people -21%) and cyclists from 28 to 24 (4 people -16%). Confidence intervals are very wide with such small numbers of people. Across the city the reductions were -29% (673 to 444) for pedestrians and -42% (475 to 274) for cyclists.  

In interpreting the significance of comparisons between changes before and after in the 20mph areas and citywide we note that Phase 1 comprised 20% of city roads. Before pedestrian casualties there were just 8% of the city total (53/673), so not representative in risk (ie Phase 1 is already 2 and a half times safer than average roads for pedestrians).  Before cyclist casualties in Phase 1 were 6% of the total for the city (28/475) i.e. over 3 times safer than average roads. This skews direct comparisons of percentage casualty reductions. Phase 1 roads already had far below average casualty rates due to characteristics present before the 20mph limit.

Other councils are reporting good reductions on 20mph limits in all casualty types – for instance Calderdale -22%.  Results in Manchester need further detailed analysis on what happened where to understand whether they are of any significance and what could be done to raise driver compliance. 

Speeds fell an average of -0.7mph. Experience of the introduction of 20mph limits in other cities is similar. But average speeds are a crude measure of “pace” and “feel” of a road which is so important to active travel and liveability.

Of transport funds of £800k, £113k was allocated to complete the signage for phase 2. It was resolved that the remaining £687k could be for other appropriate road safety schemes for speed management.

Rod King MBE, Director of 20’s Plenty for Us said: “We welcome the report and the commitment within it for the ongoing program of 20mph limit implementations. Further analysis will enable them to be optimized and enhance their implementation.  Manchester has a strong commitment to making its streets healthier, safer and better for all.  Like most of our iconic cities it sees 20mph limits at the heart of those plans. It is incorrect to use early, small number datasets which are not statistically valid and compare them to whole city totals. Manchester’s challenge is to make the rest of its 30mph roads as safe as its 20mph roads. Setting 20mph limits with appropriate public engagement and enforcement will be key to that.”

Open PDF

Also see our blog on Lessons to learn from Manchester

Showing 2 reactions

Please check your e-mail for a link to activate your account.
  • Rod King
    commented 2017-03-11 13:02:41 +0000
    Hi Gina

    When you put in wide-area 20mph limits you include many roads where the average speeds are already low. You do this for consistency and because it would be ridiculous to maintain a 30 limit on such roads when “faster” roads had a 20 limit. Reductions tend to be proportional to the previous average, with higher ave speed roads achieving the highest reductions. Hence the inclusion of low speeds roads where you will see minimal or zero reduction tends to dilute the overall average reduction.

    On most 20mph wide-area implementations you get approx 1-2mph overall average reduction in speed. But this is skewed with faster roads achieving up to 6mph reduction. Whilst this is quite different from an isolated and physically calmed scheme, such schemes only reduces speed on that small section of road and endorses 30mph on the rest of the road network. You can see our comparison at

    You are correct in considering that authority-wide 20mph limits are far more about setting a social consensus rather than traffic management. The crude sort of analysis conducted by Manchester City Council makes no allowance for this and is critically flawed. See my blog on this at
  • Gina Dowding
    commented 2017-03-11 11:52:19 +0000
    Please can you explain this statement further ?
    “Speeds fell an average of -0.7mph. Experience of the introduction of 20mph limits in other cities is similar. But average speeds are a crude measure of “pace” and “feel” of a road which is so important to active travel and liveability.”

    I am completely behind the move to 20mph but are average reductions in speed only really less that 1mph ?

    Also – if other streets are becoming ‘safer’ in non 20 mph areas – can that be linked to an effect on over all carefulness of drivers?