Written evidence submitted by Dr Tom Cohen


I am Senior Lecturer in Transport at University of Westminster and part of its Active Travel Academy.  I chair a successful series of meetings called Cycling@Tea-time; we have just launched its sibling, Walking@Tea-time.  My research centres on the formation, appraisal and evaluation of transport policy.  As part of this, I have investigated transport innovation, particularly automated vehicles, and policy disruptions such as that caused by Uber.

I consider myself to be genuinely agnostic about e-scooters, which may make me a rarity.


The title of this inquiry[1] is illuminating for two reasons:

I shall address these points in turn, then briefly discuss the role of the COVID-19 pandemic, and finish by offering some recommendations.

Decision making under uncertainty

We of course want to know whether e-scooters would be a good or a bad thing.  This is made complicated by the number of relevant considerations: road safety, accessibility, congestion, environmental impact, public health (in no particular order).  It is very probable that, like other novel forms of transport, e-scooters would turn out to be a mixed bag – good on some fronts and less so on others.  But the issue at hand is that we know very little about their likely impact.

A great deal turns on the issue of mode shift: how were our putative e-scooter users travelling before they switched?  If they were walking, the move to e-scooters would be generally regrettable.  If they were driving, we would probably regard the choice of e-scooters as positive.  Given the lack of UK data, we naturally look abroad, but other locations can only tell us so much.  I would, for example, be very wary of assuming that US experience would be replicated here, especially as so much riding there is taking place on the (wide) footways of cities built during the automobile age. Even locations with topographies more like those of UK cities could prove unreliable reference points because of differences of culture, road law and so forth.  The one thing I would say with some confidence is that e-scooters could be expected to be popular.

This degree of uncertainty when formulating policy is nothing new.  But it does require us to accept that deterministic forecasting tools very commonly used in transport are not what is needed here.  We instead need an approach to planning based more on scenarios.  Let us consider at least two situations, one based on mass migration from human power (mainly walking), and another on shift from the car.  Is there a course of action that is robust against both of those?

The role of innovation

This is not the place to launch a debate on the true worth of innovation.  It is, though, important to identify the role that the concept of innovation has played in the discussion of e-scooters up to now.  E-scooters have tended to be characterised as innovative and their champions are certainly content to support this portrayal.  Whether they actually are is not the issue.  But the tendency to celebrate so-called innovation can cloud judgement.  Two technologies may have exactly the same practical strengths and weaknesses but, if one is considered innovative, it will tend to be preferred.  I therefore simply claim that we must try to set questions of innovation completely aside; let us instead appraise this technology on its merits in transport and other relevant terms.  If something innovative makes the world a better place, it does not do so because it is innovative, after all.

The COVID-19 pandemic

Whilst the call for evidence does not mention the pandemic, its relevance to the inquiry is obvious, particularly given the amount of current debate about how the transport system will function as restrictions are relaxed.  In a world where we are asked to avoid public transport, where many trips are too far to walk (and not everyone wishes to or is able to cycle), and where a mass return to the car is untenable, it is natural to look favourably on e-scooters.  Hence the expansion and acceleration of trials in UK towns and cities.

From a regulatory perspective, the exceptional and (we assume) temporary condition of the transport system during the pandemic requires us to differentiate clearly between what may be appropriate now and what may be appropriate when restrictions are lifted.  In particular, if the use of e-scooters is enabled now (for reasons set out above), it is important that this action can be readily reversed if the conclusion is that they are not appropriate in a post-COVID environment.  And the challenge for this inquiry (and policy makers more generally) is to try to think ahead to a post-COVID time.

Temporary or experimental regulation and/or the use of sunset clauses can provide governments with the means to return to the status quo or an acceptable variant of it.  These could prove valuable, given the difficulty generally experienced when attempting to reverse policy.


I have above recommended:

In addition, it is important to emphasise the role of properly planned and resourced evaluation.  The trials being set up in the UK must be seen as trials and a concerted effort made to learn from them.  They should not be seen as a formality preceding legalisation.  Central government should take responsibility for the evaluations and plan them in collaboration with local partners hosting the trials.  It should not fall to the partner authorities either to design or to finance the evaluations.

A wider point relates to the policy context in which e-scooters may be legalised.  If we wish the majority of trips on e-scooters to transfer from private motorised transport, policies supporting this should be put in place.  E-scooters can be seen as a viable alternative to the car for many journeys so, if provision is sufficient and pricing reasonable, their introduction could legitimately be combined with measures to discourage car use.

My final recommendation concerns the highway environment.  On the assumption that e-scooters would not be allowed to travel on footways, the question then becomes how to accommodate this new mode safely on carriageways which are already shared by a range of user types, often to and beyond capacityIf, for the sake of argument, it is decided that e-scooters should have access to cycle lanes, this would have the effect, ceteris paribus, of decreasing the level of service for cycling as a mode, making it a less attractive option.  If, on the other hand, e-scooters were expected to share lanes designed for general traffic, this would probably affect negatively conditions for motorised vehicles.  The issue of road safety permeates all of this and I am not advocating one course of action over another – any decision concerning the right to use sections of the carriageway would need to be made with regard to road safety as well as the range of other considerations mentioned above.  But the wider point echoes the previous paragraph: it would be a mistake to introduce regulation for e-scooters as if they would operate in a vacuum.  If a new category of vehicle is to be allowed onto crowded carriageways, this requires a coherent supporting strategy that addresses how overall demand is both catered for and managed.

June 2020


[1]              e-scooters: pavement nuisance or transport innovation?