In brief 

• OIA reveals no NZ statistical data to substantiate claims that petrol cars are more likely to spontaneously catch fire than EVs. 
• NZTA and FENZ lack of comprehensive data makes claims EVs are less likely to catch fire “fake news”.
• Data from FENZ suggests more than a fifth of all vehicles randomly bursting into flames involve “EVs” as defined by FENZ.
• EV fires’ dangerous unpredictability also raises concerns.

Click for Part 1, Part 2, Part 3, Part 5

Fact-checking an EV fire myth

It’s often said in the media that petrol cars are far more likely to catch fire than EVs. It might not be true.

NZTA Waka Kotahi principal engineer Don Hutchison told a colleague asking about the “standard response” on EV safety, 

“We don’t have much to add other than we understand that the Fire Service statistics show that a battery fire is a lot less likely than a petrol or diesel fire.”

However, in an Official Information Act release to Centrist, FENZ admitted it has never collected that statistical information.

We posed a series of questions designed to work out how common EV fires are in NZ and, more importantly, whether they are more likely to spontaneously combust than ICE cars. Here are some of the questions:

3. The total number of motor vehicle fires attended in the last 24 month period for which complete data is available
4. A breakdown into how many of those fires featured lithium-powered transport (EVs, electric scooters)
5. The total number of house or structure fires where a lithium battery or charger was deemed to be the proximate cause
6. A breakdown into how many house or structure fires were attended in the same period and deemed to be caused by a parked ICE vehicle switched off (eg, not an ICE car crashing into a house resulting in fire)
7. How many incidents of any kind were deemed to be caused by spontaneous combustion (ie, no other known contributing factor) of a switched off or charging lithium-powered transport, or lithium-powered appliance or device (eg laptops, phones, tablets, cordless electric power tools etc)
8. How many incidents of any kind were deemed to be caused by spontaneous combustion of a switched off ICE vehicle or power tool (eg, chainsaw, line trimmer, lawn mower etc)

FENZ spokesman Aidan Saunders replied, confirming “we do not record or categorise incidents for questions 4-8.”

Claims that EVs are less likely than ICE cars to combust are “fake news”

You would think the Fire Service would want to know exactly what its risk exposure was to EV and other lithium fires. Yet, based on their OIA response nearly two years after Jacinda Ardern and James Shaw announced NZ was aiming for EV dominance of car sales by 2035, FENZ has never asked the hard questions by standardising its incident report checklist.

If the Fire Service is not collecting data on EV fires, that means they can’t actually prove that EVs are less likely to catch fire. It’s “fake news”.

Saunders gave us some data, but admitted that FENZ is “unable to confirm the data provided to you is an accurate total of the number of incidents that have occurred.”

So what data did they give us? The total number of car (mobile property) fires from any cause, Q3, is reliable data, and the answer is 1961 car fires in 2022, 2414 car fires in 2023, and 216 car fires in the first month of 2024.

Of those mobile property fires, which FENZ says will also include e-bikes and electric scooters because FENZ data collection doesn’t separate out battery EVs from hybrids or scooters, only 35 vehicles spontaneously combusted over the past two years.

Two of those were definitely EVs of some kind, but the type of vehicle is not recorded for the other 33 because FENZ didn’t  check.

However, we did.

A search of news articles reveals:

2022

• Mercedes hybrid fire
• Nelson EV fire
• Wellington EV fire

2023

• Pakuranga EV fire
• Wellington e-scooter fire
• Tesla fire Auckland Harbour Bridge
• Dunedin EV fire
• Wellington e-bike fire

FENZ data suggests more than a fifth of all vehicles bursting into flames involve EVs

So, if the Fire Service recorded 35 spontaneously combusting vehicles in two years, and we can find eight incidents over the same period that meet the FENZ definition of an electric vehicle, that suggests 23% of all vehicles bursting into flame without cause (ie, no crash involved) were EVs, based on the only data FENZ has.

And given that EVs only make up 2.38% of the New Zealand vehicle fleet, this rough number-crunching of Fire Service data indicates EVs may be 10x more likely (if you use FENZ EV definition, or 7x if you restrict it only to cars) to randomly catch fire than petrol cars.

Why is this important? The key word is randomly. All of us understand why an ICE car might catch fire in a crash. We also understand why poor maintenance or electrical faults might cause fires (something that affects EVs too).

What we find disturbing about randomness is the unpredictability. An EV fire on the Harbour Bridge or Waterview Tunnel at 3am probably won’t be a problem, but if fire breaks out in bumper to bumper rush hour traffic at a standstill, with no way for other motorists to escape, it’s a whole different scenario. An EV can be fully alight in just two minutes. 

Flame jets of 1.5 metres have been measured shooting out of batteries, easily capable of igniting the tyres or fuel tanks of nearby vehicles at a standstill. The world’s road tunnels have been tested to see how they cope with “an” EV fire under ideal conditions, but they haven’t been tested to see whether the tunnel can survive a 20-car conflagration.

A “dumb” question raised

“This is probably a dumb question,” NZTA’s Peter Brown asked a colleague when last year’s Tesla fire on the Harbour Bridge made the news, “but have we developed any protocols around dealing with EV fires for AHB and tunnels?”