A significant gap in the planning and delivery of the national public charging infrastructure network has always been knowing exactly where the Battery Electric Vehicle (BEV)s are currently located to help match supply with demand.

Fortunately, on May 20^th, this position significantly improved with the release of the snappily titled “Electric Vehicle Charging Device Grant Scheme Statistics”^[1].  This dataset details where the residential chargers are.  As just about everyone who gets a BEV applies for a grant for their home charger, this must be the golden record?  Not quite, but it is a big step forward.

Locating where BEVs are is hard because there isn’t really a logical record that can be used.  A car or van isn’t necessarily owned by the person who drives it or even owned by the organisation that provides the car to the driver.  For example, a company may provide a car to an employee but that car may be leased, and therefore owned, by a finance business – so where should the recorded address be – the driver, the company or the finance business? The answer is a rather messy mixture of all.  This confusion means that the standard DfT data for BEVs (VEH0134)^[2] is hopelessly skewed.

Instead, wouldn’t it be great if we had a data set that recorded the location of a specific BEV asset?  This is what this new dataset does. The great majority of BEVs today are bought for drivers who have access to off-street parking that enables them to charge the BEV overnight using a residential charger.  The government is very generous in its Electric Vehicle Homecharge Scheme (EVHS) grant for residential chargers, so having a EVHS charger installed is part and parcel of most BEV drivers experience. This new dataset records the location of all EVHS grants, and the previous Domestic Recharge Scheme (DRS) grant, by council or postcode district and so is an excellent record of where BEVs are actually parked at the end of the day.

As there are 172k charger locations and 208k recorded EVs (BEVs) in GB, then at first glance this data suggests an 82% match, which is far better than any other data set – but caution is needed.

There are a number of local variations that we should be aware of:

Trials – In the early days, a number of large trials were carried out where drivers were given or received heavily subsidised chargers and cars.  It is not clear what their behaviour was post trial so we should be suspicious of the large early spikes in DRS data, particularly around Sunderland and Nottingham.

PHEVs – Many residential chargers were and are installed for Plug in Hybrid Electric Vehicles (PHEVs).  This will overstate the number of BEVs in an area.

Eligible chargers – Not all chargers are eligible for the grant, particularly the Tesla charger, and as the Tesla is the bestselling BEV this could cause a distortion.

Moves – The EVHS grant is available once, and so when someone moves the data “stays” at their previous address.  Sorry Worcester, you have me now but don’t know it.

ICE Swaps – Not all EV drivers keep their BEVs, some return to petrol cars but the charger record stays with the house.

On-Street parkers – Nationally, today, the vast majority of BEV drivers will have a drive, but in certain dense urban environments this trend is reversed.  In Tower Hamlets only 7.3%^[3] of households have a drive so BEV drivers in this area will nearly all be parking on the street.

What all these caveats mean is that this residential charging data is great but needs to be interpreted. It is not the golden record but is a real step forward for councils to use to understand their environment.

To understand how to apply this data, let’s consider the detailed chart below:

The chart describes the BEV and charger counts provided by the different data sources.  Each coloured dots represent a different way of trying to establish EV counts:

• Grey refers to taking the average of BEVs as a part of the national fleet (0.58%) and applying it to the total number of all vehicles in each council. This will overstate areas of low uptake and understate areas of high uptake.
• Blue refers to taking the DfT data on BEVS and normalising it to the Personal registration data. This removes many of the excessive variances of the DfT registration data but is still significantly impacted by leasing contracts recording EVs at the leasing HQ, not at the drivers residence
• Yellow refers to the unfiltered DfT registration data, this is very misleading but is included for records only
• Purple is the new EVHS and DRS residential charger data

The overall observation is that there isn’t really a good quality correlation between the different data sets but some broad observations can be made.

The National Average number and the residential charger data is consistently quite close.  This suggest that the national average number is not a bad proxy for some councils to use.

In London, the lack of off-street parking is coming through as a gap between the residential charger counts and the different estimates for BEVs counts.  This suggests that in London care should be taken when using the residential charger counts.

In the Counties and Districts, the residential charger counts are close to, but behind the national average and adjusted DfT data by 10-20%.  This suggests that the residential data should be used and considered but with local reference to the level of on-street parking.

In the Metropolitan Districts the adjusted DfT data is very low which either suggests a high company car take up or perhaps more likely a heavy data skew from a few leasing HQs, particularly Lex Autolease in Stockport.

In conclusion, this data set is a really helpful addition in the search for local EV numbers and the DfT should be applauded for releasing it.  However, we would suggest that it should always be used in conjunction with On-Street parking data, so that its relevance can be most accurately assessed.

What will we do:  We will shortly be adding this data to our On-Street Charging Map, so that you can understand all the relevant data in one place.