Something remarkable happened in the last five months of 2025: a massive reversal in outcomes from the first seven months of the year meant New Zealand recorded its best ever road safety result.
Or perhaps “least bad” is the better phrase. In total, 273 people tragically and unnecessarily lost their lives on our roads last year. And while that is still way too high – especially compared to many of our peer nations – it is arguably the lowest total we’ve ever seen, since records began.
That we achieved this at all is remarkable, given where things were in the middle of the year. In 2024, 292 people died on our roads – a big improvement on previous years, and the first time since 2013 there had been fewer than 300 deaths.
However, by the end of July 2025, the 12-month rolling total had jumped back up to 324, meaning 2025 wasn’t looking great.
Then in August, just 14 people died – the lowest ever result for an August – and that was followed by just 8 people in September: not only the lowest September result on record, but the lowest for any month. It was even lower than April 2020, when almost everyone was at home during the nationwide COVID Level 4 emergency.
The numbers were a bit higher in October and November, but were still some of the lowest on record for each of those respective months. Then, another lowest ever result in December when just 17 people died. This was a huge change from the 41 people who died on our roads in December 2024, and less than half of the average December for the previous 10 years.
There will be a few reasons for the outcomes we’re seeing. Likely the biggest of these is the substantial increase in enforcement by police. For example, since mid-2022, the number of speeding tickets issued by officers has increased by around 50%, in line with both the previous government and the current one requiring more from police on road safety.
It is also likely we’re starting to see the benefits of the investments made by the previous government into things like median barriers and other, smaller highway upgrades.
Adjusting for population and distance traveled
The total of 273 deaths last year is still higher than the 253 people who died in 2013 (a previous record low) – however, New Zealand is also a lot bigger than it was then, by nearly 900,000 people.
The number of deaths per 100,000 people is a common metric used internationally to compare the road safety performance of different areas, and on that basis 2025 was arguably our best ever year. Our population increase means that last year saw 5.1 deaths per 100,000 people, below the 5.5 of 2024 and 5.7 of 2013. You have to go all the way back to 1922 and 1923 to get a lower result – and cars certainly weren’t as common in society then as they are now.
Even so, it is worth noting that this figure is still much higher than many of the countries we like to compare ourselves to. In 2025, Norway – with a population only slightly higher than New Zealand and with a more difficult geography and climate – had its highest number of road deaths in years, with 111 people dying, which is just less than 2 per 100,000 people.
Many other countries in Europe are in the 2-4 range. And, for our immediate neighbours across the Tasman, the Australian state of Victoria was just over 4 for 2024, with the whole of Australia at 4.78 deaths per 100,000 people in 2024 (the most recent available data).
There is also the suggestion that the state of our economy may be playing a part, resulting in fewer people travelling. However, this isn’t borne out by the numbers. In the year to the end of September 2025, vehicles travelled 49.92 billion km on New Zealand’s roads (VKT). That is the highest it’s ever been, up from 49.2 billion the year before and 40.3 billion in 2013.
Both the per capita and per km travelled metrics are shown below. As you can see, they follow a very similar path. Note that both the population and VKT data are only available up until September 2024, so I’ve extrapolated the trend for December.
Who is it that’s dying on our roads?
These charts are based on some quick breakdowns by the Ministry of Transport, and show where the change in the last few years has come from.
By Mode
As you can see below, deaths in 2025 (purple) were down for most of the ways people move around on our roads.
The 2025 numbers are yet to be finalised, but so far it’s great to see a continuing reduction in pedestrian deaths, which is also a new record low – although that is negated somewhat by the slight increase in deaths of people on bikes. Motorcyclists and passengers in cars are where the biggest reductions occurred. And by far the highest number of deaths are those who are at the wheel of a vehicle.
By Age and Gender
The MOT data for road deaths by age uses inconsistent age banding, which distorts the picture somewhat. However, on a per capita basis, it’s clear that the youngest drivers (15-24) have the most deaths per 100,000 people, compared to other age groups.
Despite the total coming down significantly in the last few years, still more than twice as many men are dying on our roads compared to women.
By Road Type and Speed
We can see that deaths decreased significantly on state highways last year, while they notably increased again on local roads after a few years of reduction.
Looking at speed limits, the biggest decrease came on roads with 100km/h speed limits, which also happens to be where most deaths occur. This reduction is likely a reflection of the police being required to focus enforcement more on the open road.
Meanwhile, almost all of the increase in road deaths on local roads was in areas with 80 or 100km/h limits – and you have to wonder how many of these crashes may have been impacted by Simeon Brown’s increased speed limit mandates.
In July 2025, for example, after speeds were reverted from 80 to 100, there was a death on the road between Ōtaki and Levin, with one local noting: “When the speed limit went down to 80kms we didn’t have any fatalities on our road. There wasn’t any significant crashes on our road and then – as soon as it went up – there’s been two crashes today that I know of.” (It was one of a cluster of fatalities on affected roads that we noted at the time.)
A look at the regions
Finally, there were 40 deaths on Auckland’s roads in 2025. That’s up from 36 deaths across 2024.
And across all regions, there were also notable increases in the Waikato and Hawkes Bay. Impressively, there were no road deaths at all in Gisborne last year.
Greater Auckland’s work is made possible by generous donations from our readers and fans.
We’re now a registered charity, so your donations are tax-deductible. If you’d like to support our work you can join our circle of supporters here.
You can also follow us on pretty much every social channel, or find us on Substack.
Processing...
You say “it’s clear that young drivers (15-24) have the most deaths per 100,000 people, compared to other age groups”, but then show a graph title Road Deaths by age, that shows the opposite.
The over 60s age band has more than double younger drivers (if we define that as people under 25) and significantly more than 25-39 year olds or 40-59 year olds.
This older age group also shows a significant increase over last year, whereas other drivers show a decrease or minimal change.
An additional age band might be useful as 60 years old and upwards spans a growing proportion of the population. This might simply reflect older drivers generally in a way not captured in the ‘per 100,000 people’ scale.
“the most deaths per 100,000 people” – the graph is showing absolute numbers, not per-capita numbers. There are roughly 700,000 people in NZ aged 15-24 vs about 1.1 million aged over 60 – so the per capita rate is higher for the younger group (noting that their numbers are split across two bars)
About 55% on 100 km/h roads.
Keep rolling out those median and side barriers.
Deaths/100k people by region is interesting. Based on the graph and a google search of populations:
Auckland: 2.2
Canterbury: 4.6
Waikato: 12.4
Anecdotally, the driving culture in the Waikato is dreadful. Red-necked bogans-with-bank, roaring around in lifted 4x4s festooned with bullbars and racks of spotlights. The only time that I’ve ever seen a car stuffed into the CANOPY of a tree was in the Waikato.
Crappy driving combined with a considerable amount of through-traffic, I am not remotely surprised that the Waikato has such a high death rate.
The ongoing non-delivery of SH2 improvements from Pokeno to Mangatarata will contribute to the fatalities – this is a busy and unsafe road. https://www.nzta.govt.nz/projects/pokeno-to-mangatarata
From what I see here the driving isn’t significantly worse than any other bogan-filled region. The big difference is that Waikato covers a massive area (Pokeno to Taupo) and has huge amounts of through traffic, as well as people who come into the region for work every day. In my view this makes comparisons to other areas hard, while also highlighting the fact it should be first in line for improvements like median barriers etc.
Agreed, Adam. Doing it on a regional population basis is a bit meaningless when so much of the traffic in Waikato is driving through from Auckland, Bay of Plenty and beyond. The people actually living in the region are a relatively small part of the problem.
Also not surprising that Auckland’s rate is so low – it’s predominantly urban and (as we saw in the other graph) most deaths occur in rural areas.
Interesting to see how high the death toll is for Waikato and Manawatu- two areas which have closely settled farming regions which can mean lots of young males in hoon cars and utes.
Areas strong on “car culture”
Waikato and Manawatu are both areas strong on state highways running across the region. Deaths occurring here is not the same as deaths of people from here.
They are also the two regions that have more traffic from outside the region passing through then than any other region.
Stats on deaths are always available earlier than for serious injuries. This makes death stats more “newsy” than serious injuries that have a similar aggregated effect on people (except that dead is final for that person).
The difference between death and serious injury as the outcome of a crash is a very fine margin, so we really do need to look at DSI stats to get a full picture of trees in harm.
Good point. We should really have a general category for “serious accidents” regardless of the injuries. New cars are pretty safe so it can be easy to get complacent and think that are roads are improving more than they actually are. Death and injuries aren’t the only impact of serious accidents too – even ones with no injuries make a huge mess and cause major congestion. It would be good to have a more nuanced picture of what is happening on the highways.
DSI is about people and familiar, health services, employment etc.
Another topic is incident effect, which can have a big economic impact without necessarily any physical injury. For example, An incident on Auckland Harbour Bridge at peak time caused complete gridlock across Kaipātiki Board area for hours.
Would love to see a deeper dive into cyclists and pedestrians.
Cause of accident, where did it occur. If someone was found at fault were they prosecuted?
Yep, the focus on a few hundred fatalities doesn’t provide huge insights into the nature of more serious crashes. From what I can see, we had just over 2400 serious injuries on our roads in 2025 as well, which was also down slightly on previous years. There were also over 10,000 minor injuries last year (again, down on previous years). Analysing them as well is likely to give us a much clearer picture of the key problem areas to target.
(was meant to be in reply to adam’s comment above…)
Further analysis is always welcome, particularly the sort of work you’ve been doing, Glen. But in some ways it is largely irrelevant if the politicians make a culture war out of the obvious conclusions from such analysis: lowering speed limits, infrastructure for walking and biking, traffic calming, low traffic neighbourhoods, putting streets on diets, reducing VKT, refocusing vehicle safety onto their impacts on active travellers.
How is the data worked out for streets that were 50, became 30 briefly, and returned to 50, or that had other speed limit changes? Could the MoT split the data into road types described by how their speed limits changed?
The increase in number of deaths on 30 km/hr streets is notable, given length of streets at this speed limit has decreased.
Not really, because most streets didn’t revert back to higher speeds until the middle of the year (and in most cities, very few streets were reversed). Even then, it seems that Aucklanders pushed back on this and stayed travelling slow anyway.
Many of the 30km/h speed limits were introduced in 2023-24 and even early 2025, so the perception that there were more deaths on these streets is simply because there were now many more streets with these limits (probably also explains the drop on 50km/h streets – there were fewer of these streets…).
Yeah. The timing matters. Makes it tricky.
“Aucklanders pushed back on this and stayed travelling slow anyway”
Is that something you’ve been able to analyse, Glen? Anecdotally, I’d describe our speeds as having distinct parts: some people are more confident in travelling safely, others are more confident in travelling recklessly. In the process, poor driving behaviours like cutting corners and driving along footpaths has only increased. I would posit that Simeon injected extra carelessness into car culture.
The distribution of deaths on residential local streets that were covered by the 30 k speed limits would need all 5 years before statistical evidence would show up. Many of those streets simply confirmed the speed that people drove anyway. Others had some traffic calming added, so speed dropped and has stayed low.
30 k limit effectively normalises safe speed for that type of street – the point that Heidi rightly stresses.
Low speeds prevent deaths and injuries that would happen on those streets when people enjoy the freedom to walk bike and even play around their homes and schools.