Sam Warburton welcomes public debate over the road toll but bemoans that it has taken horrific accidents and many deaths to trigger it

Picture: Radio NZ

By Sam Warburton*

Two weeks ago, and as part of an ongoing series about transport planning, investment and outcomes, I wrote about the road toll.

In that period a further 15 car occupants have died, including four people north of Taupō this past Tuesday.

That accident occurred at the intersection of Tutukau Road (running East to West) and State Highway 1 (running North to South) as marked on the map below.

Waikato councils put forward that section of State Highway 1 between Wairākei and Ātiamuri as a candidate for safety improvements. In 2015, the NZ Transport Agency tagged the project as ‘probable’, meaning that the project would likely progress if funding was available.

Funding was due to begin in 2015/16 for preparing a business case, and continue through to 2017/18 with construction. It appears that the project is yet to receive any funding.

We can’t say what factors contributed to this crash. Nor can we say whether safety improvements on that stretch of road would prevent similar accidents. We can, however, say that these are not hypothetical lives. And these are not hypothetical projects.

Safety is an important aspect of transport for many travellers. The country should be debating whether we have the mix right between safety and other features of transport we care about including travel times, resilience, access and choice.

If we, as communities and as users of transport, are unhappy with what we’re getting, we should be asking whether government and its agencies should be the ones deciding for us.

Today I want to respond to some of the coverage of the road toll over the past two weeks and to tease out these issues more.

The road toll is definitely increasing

Several commentators questioned whether the road toll is increasing given that population has increased since 2013.

The NZ Transport Agency was among these, telling Radio New Zealand: “There’s definitely a trend up but equally there are more people driving. The amount of kilometres that people are driving in their cars has risen by about 10%, 15%.”

This view is mistaken.

As I noted in my previous article, the fatality rate (deaths per kilometre driven across New Zealand) of car occupants is 41% higher in 2017 than it was in 2013.

Another way of looking at this is that, while New Zealand is driving 15% more as a country than in 2013, deaths are up 60%.

Had we kept fatality rates at 2013 levels, there would be about 80 fewer deaths than we expect there to be in 2017, and over 110 fewer since the end of 2013.

These changes are, in technical terms, statistically significant. We can have greater than 99% confidence that the road toll is up.

Some regions are faring worse

While the long historical downward trend in the road toll was apparent across almost all regions, the increase of the past few years is not. Increased safety risks appear to be concentrated in the Upper North Island regions of Waikato, Bay of Plenty, Auckland and Northland and the lower South Island regions of Otago and Southland.

The chart below shows those regions where the risk appears to have increased since the national-low road toll in 2013.

Why is the road toll increasing?

I don’t know. And it looks like no one does.

Commentators have suggested plenty of plausible factors (such as distractions from smartphones and reduced focus on fixing dangerous roads), as well as some less plausible factors.

The Police have put it down to driver behaviour. They’ve ruled out a lack of funding as a factor.

The Police’s position will come as a surprise to the Police who, in February, told The Otago Daily Times that funding issues were why 12 New Zealand regions have been without fixed speed cameras for 18 months.

To really understand what’s going on requires robust statistical analysis.

The Ministry of Transport has undertaken at least three studies into the road toll since 2013 for various purposes. Methodological issues mean that these studies cannot be used to explain the increase. One study says the biggest factors affecting the road toll are petrol prices and wages. Petrol prices and wages affect how much we drive, less so the chance of dying once we’re on the road. The solution to the road toll is not higher petrol prices or lower wages.

According to people I’ve spoken to recently, the Ministry commissioned a fourth study that was due to be completed in March 2017. It is yet to be released.

Those same correspondents are looking at undertaking their own research. I look forward to the contribution this will make to the public debate.

Could New Zealand be doing more?

Yes. Proposed improvements between Wairākei and Ātiamuri in Taupō show there are other projects out there that could be pursued.

The question is whether these interventions and investments are worth it relative to the other transport choices we have, such as transport to support new housing developments.

In interviews about the road toll, the NZ Transport Agency has talked up the safety benefits of the Roads of National Significance. However, as I noted in my previous article, these seven roads are relatively lightly travelled compared to the average $1.6 billion each cost. And $1.6 billion could buy a lot of road median barriers and road safety ads.

Where to from here?

The public debate that has erupted over the past two weeks is welcome, though it is sad that it took horrific accidents and many lives to trigger it.

What we now need is a structured and responsive way to address these issues. Different safety interventions have different impacts, and safety isn’t the only aspect of transport we care about. Focussing on one aspect at a time risks swinging too far from other important objectives.

In my next piece for Interest.co.nz, I’ll set out current planning processes and how the public can be involved. I’ll also start to look at how that process could be reformed.

Road deaths

Select chart tabs »
The 'Monthly count' chart will be drawn here.
Loading...
Source: NZTA

*Sam Warburton is a research fellow at the New Zealand Initiative, which provides a fortnightly column for interest.co.nz.

We welcome your help to improve our coverage of this issue. Any examples or experiences to relate? Any links to other news, data or research to shed more light on this? Any insight or views on what might happen next or what should happen next? Any errors to correct?

We welcome your comments below. If you are not already registered, please register to comment or click on the "Register" link below a comment.

Remember we welcome robust, respectful and insightful debate. We don't welcome abusive or defamatory comments and will de-register those repeatedly making such comments. Our current Comment policy is here.

66 Comments

Hmmm, the rental vehicle market is pumping... https://www.nzx.com/markets/NZSX/securities/THL?icharts=true

More vehicles probably means more driving. But as I have written in both my pieces, the road toll is up way above and beyond the increase in driving.

We're driving 15% more than in 2013 but dying 60% more.

Just a note, Sam. Thought you had a great Lions tour. Well done.

You know, I never get tired of that. :|

Yeah...it was my first time and you can rest assured I'll never do it again :)

I think the sentiment is much the same for Zachary Smith :D

Is there a disproportionate % of rentals involved in road fatalities (per km travelled)?

It's beyond my skills to analyse all the causes. But it's not beyond others' skills. Government agencies should track down some of those statisticians/econometricians and get them on the job.

If you spread it throughout the country it does not look too bad. Take somewhere like Central Otago or Southland or the West Coast and it is nothing short of alarming. One of the bigger problems there is roads that are seemingly empty, giving a driver more used to driving on the right, no "guide" from being among other traffic.

I doubt the amount of road has increased 15%. So therefore you have more cars per km of road and all the problems that go with it like longer queues at intersections, trains of frustrated drivers held up behind caravans, much tighter gaps when merging into arterial roads.

It would be interesting to see the trend broken down by the different crash scenarios. How many are at intersections vs open road? How many involve more than one vehicle etc.

At some point adding more laws, rules and limits becomes detrimental. As was the case when they introduced the no tolerance policy for going over 100km recently. Some theorised that the road toll increased because people were paying more attention to not breaking the speed limit than actually focusing on driving their car. There's a good case for removing laws rather than adding them - http://www.dailymail.co.uk/news/article-1028740/Accident-free-zone-The-G...

Agreed, trying to remove STUPID from the drivers seat through rules and regulation will never be entirely successful. Some of the accidents are due to wilful stupidity, while others are the result of momentary lapses in attention. Both are human factors and neither can be effectively ruled against.

Unfortunately, not all fatal accidents caused by willful stupidity result in the demise of the willfully stupid.

A rather obvious factor is the 'magic thinking' generated by the existence of so many black-box doo-dahs in vehicles now, compared to even a few years ago: A short list:

  • Predictive braking
  • lane departure warnings/action
  • chassis dynamic adjustments carried out several times per second

These and other forms of what to most folks is Magic, engender a feeling that 'I'm safe, the car will look after me regardless".

So actions like expecting traction control to work when hydroplaning or on black ice, taking corners really fast, not reading corners which then tighten up (a skill burnt in to any surviving motorcyclist...), and getting distracted by touch-screens, children, partners, pax, entertainment systems (it's a very long list) then go on to cause the Unexpected - the proverbial 'oh, crap' moment.

Maybe it's because crashes are random and using years as measures for averages are not applicable?
Maybe it's because of the large amount of rental vehicles and tourists using cars, not understanding or knowing the roads which contribute?
Maybe it's because our roads are arse?
Maybe it's maybelline

There is an element of randomness to crashes, but the trend is definitely up. The method I used is standard for such issues. We can be more than 99% confident that the road toll is up.

As everybody knows It is pointless using deaths as the stat for road the road toll. Just saying we have got better or worse on that score year by year. That is what is thrown at the public to digest. An accident that kills one and cripples say four there is one stat. Needs to be a more graphic scale representing all the horrors surely.

Serious injuries have the same pattern.

Sure it might be the same pattern but your article quite rightly seeks to stimulate public interest and debate. So a few years ago or so when we had a drop in road deaths per annum, there was some trumpeting about the improvement in our driving. That in itself illustrates a large degree of how public perception can be misled.Obviously it is not as simple as this but if those sort of stark & limited statistics are bandied about without including for instance that serious tourist bus accident in Canterbury a couple of years ago or so, where nobody died but about 15 or so were left with crippling injuries, then it is providing neither the correct or serious message. Appreciate this is not a major point but if only, to take it is an as an angle, then it fails to deliver the full picture, and it is part of the problem of insufficient education of the public.

Anything with a degree of randomness is going to be 'lumpy', with bare spots and clusters. Maybe the low toll of recent years was the anomaly. My statistics is rusty, but presumably there are ways to determine if it's lumpy beyond normal expectations.

Element of aging population with health and eyesight issues who are understandably reluctant to give up their licenses?

Draw a line-of-best-fit between following the peaks and a line-of-best-fit following the troughs. 2013 was not unusual for troughs.

Also a professor of statistics has been in contact with their own testing. Trend is up and it's up since 2013. Good evidence that 2011 is not the turning point.

Here is that professor. 80% chance the lowest point was around 2013/2014.
https://www.statschat.org.nz/2017/10/13/road-deaths-up/?utm_source=feedb...

up
11

It's politically incorrect, but they need to look at the drivers are who cause fatalities. While the public safety campaigns target everybody, including most of us who are already doing their best, it would be more effective to intensively target criminals, including if needed handcuffing them to their letter boxes. Acknowledging of course a fatal crash can be caused by any member of society.
1. the large number of fleeing drivers who regularly kill themselves, and unfortunately others. The latest menace was called Morrocco.
2. Who was that researcher who pointed out that 50% of fatal crashes involved somebody with a criminal conviction - but those people are only a very small % of the whole society. That research is ignored.
3. The recent research which shows us that one third of fatal crashes involve a driver with drugs in their bloodstream ( as different from alcohol.)

Some matters for research.

Of the fatalities:
How many were not wearing seat belts or in approved child restraints?
How many vehicles involved in collisions did not have warrants or registration?
How many vehicles were more than ten years old and therefore lacking modern safety enhancements?
How many drivers involved in fatalities were unlicensed or not NZ tested drivers? How many drivers involved in fatal accidents relied on driver training from other countries to get a NZ licence to drive on NZ roads?
How many fatalities were medical incidents? Is an elderly man dying in his car in a hospital car park usefully counted as a road fatality?
How many fatalities were the result of police pursuits?
How fatalities many were deliberate i.e. suicides?

These are not issues that can be dealt with by better roads, more speed cameras, or more booze buses in the suburbs. Many arise from NZ being a poor country with an aged vehicle fleet, many drivers who are not trained to drive in NZ conditions, an aging population and poor support for mental health.

Delving into these matters might lead to better targeted prevention programmes than the present breathalyse people in 50km areas where there has never been a fatality and speed traps all over the place. It may be the case that a lot of the road toll has causes that are beyond the realm of road policing or road building policy.

Vehicle age, for the most part is a crock, as is older driver age.
One of the things that is reasonably recent is the proliferation of people carrier type vehicles, which often results in far more people in any one crash than there may have been in the past.
Another is the proliferation of SUVs and other vehicles with much higher centre of gravity than regular cars.

And we are now the 3rd most obese nation. Imagine with all that weight up high....one glancing blow and watch it sway and snake across the road....

I read that and instantly this sprang to mind https://www.youtube.com/watch?v=0YtAVIGq1ko I have no idea why

We lowered the drink drive limit in 2015, but since then have seen a reduction in convictions for drink driving and increase in the road toll. This seems odd as wouldn't a reduction of the limit result in more convictions, not less. Perhaps the police have reduced enforcement efforts around drink driving.

http://www.radionz.co.nz/news/national/332589/drink-driving-numbers-plummet

Is the same thing happening in Australia?

Doesn't appear to be the same, but read this from the US

Over the last decade, new cars have gotten electronic stability control systems to prevent skids, rearview cameras to prevent fender benders and more airbags to protect occupants in collisions. Hundreds of millions of dollars have been spent on campaigns to remind the public of the dangers of drunken driving, failing to buckle up and texting while on the go.

Despite all that, more Americans are dying on roads and highways than in years, and the sudden and sharp increase has alarmed safety advocates.

The latest batch of bad news arrived Wednesday in traffic fatality estimates released by the National Safety Council, a nonprofit organization that works closely with federal auto-safety regulators. According to its estimates, 40,200 people died in accidents involving motor vehicles in 2016, a 6 percent rise from the year before.

If the estimates are confirmed, it will be the first time since 2007 that more than 40,000 people have died in motor vehicle accidents in a single year. The 2016 total comes after a 7 percent rise in 2015 and means the two-year increase — 14 percent — is the largest in more than a half a century.

The US is also facing an epidemic of drug (mainly opioid) overdose deaths - 64,000 in 2016. I just wonder how much of an overlap there is with the road toll there and perhaps here (with meth); with increasing numbers of drivers on cannabis, opioids or meth driving around in older vehicles.

https://www.nytimes.com/interactive/2017/06/05/upshot/opioid-epidemic-dr...

https://www.reuters.com/article/us-usa-drugs-crashes/drug-use-tops-booze...

Has anyone done the research into the time the driver at fault has been in New Zealand? Or the origin of the license of the driver? Police at the coal face know the answer to that one, but probably can't say it.

What would be really useful is the average speed, which could be compared across regions, and to the death rate per km travelled.

Lowering the road toll has been a campaign based on flawed assumptions for a long time. The Police released a 10 year strategic plan back in 2000 to reduce the road to to under 300. Didn't happen.

Road maintenance is down , all the money is been pumped into the RONS. Road surfaces are worst when you get off the main highways. But even SH1 between Atamuri and Taupo has been slippery as hell for the last year or so . There were a spate of bad accidents there around last Christmas, especially as it was wet.
And the weather has been very wet this year, with few (if any) weeks without rain.
I think this is where a lot of the accidents are happening , people coming off the improved motorways and expressways they ae used to , hitting the "normal " roads , and coming to grief .

Yeah I think you're right there. Living in Auckland I very rarely end up on a standard 100km/hr undivided road these days, when I do it feels really unsafe like an accident is going to happen at any time.
I think our roads and speed limits have always been terribly unsafe, but with less immigrants and less decent roads, we were all just used to it.

And those lousy road conditions are being exacerbated by the amount of rain we have had for some time. Road slumping and break up is becoming very noticeable

Yup , and the droughts before that, means the earth inder the roads is shrinking and expanding more than normal lately.

Perhaps the cell phone rule was counter-productive - it's much better to text on your lap than talk on your phone where you may be seen.

The emission rules have definitely left a huge bulge in the fleet of aging pre 2000 cars. There also was a bubble of pre-frontal impact cars imported to beat the earlier rules that the new vehicle industry had lobbied the Govt for. Then later Annette King caved into them again with the later emission rules. This made the bubble of early cars stay in the fleet a lot longer. These are huge numbers - over $1M cars which would have been mostly retired had the Govt not insisted in mucking around with the rules and the new vehicle lobby actually cared about lives instead of protection of profits.

Don't all/most phones have GPS tracking always on? If so, wouldn't it be an easy function to auto program into the devices to go to silence mode/voicemail when the phone itself is moving over a certain speed? I know I'm 'programmed' to turn off my phone when I go into a classroom or a meeting - but never do it when I get into the car.

And then the other question I have often wondered is about GPS directions/map use. I find them very disconcerting/distracting. They mean we don't have to plan trips by consulting a map before we head off.

Too much to look at and listen to aside from the road, the roads signs and the traffic.

The latest iPhones actually do this by sending a "I'm busy driving..." txt to any callers. https://support.apple.com/en-us/HT208090

Great feature - glad someone has already thought about it.

Speed alone is of little use unless you want car/bus/train passengers to switch off too :)

Some roading "improvements" are not improvements in my view.
1/ Rumble strips on the inside of the corner on left hand bends: Forces cars to be closer than necessary to oncoming traffic. I've seen many close calls. Myself: I often have my left hand wheels to the left of the rumble strip on left hand bends to keep away from oncoming trucks and their spray.
2/ Slow traffic lanes: Sometimes installed where overtaking lanes used to be like on the desert road: a recipe for frustration when the SUV in front of you feels that they are not slow. Do you overtake them on the left inside hand lane?????
3/ Unnecessary no-overtaking yellow lines on long straights.
4/ 104k speed limit: having to check the speedo all the time.
It all adds to the frustration of the driver and ups the pressure when there is a genuine overtaking opportunity.

Yes to rumble strips being counterproductive. Why encourage drivers to the middle of the road?
The margin for error on open roads at speed are small enough anyway without the increasing number of distractions that car manufacturers are putting into consoles.
Also as I mentioned in a recent thread, I hardly see traffic police these days. Are they outsourcing part of their policing to *555 ? Used to see them frequently in Taranaki through to King Country. Now would be lucky to see a traffic cop 1 in 3 trips.

Continually being bombarded by super heroes....just because they can drive a car hard and fast a 1000 times around race track, followed a series of burnouts and otherwise criminal behavior, probably doesn't help.

Maybe we should stop treating this dangerous weapon as a toy for men with small testicles?

Trucks trucks and more trucks....no or few passing lanes....trucks dont slow down and let as many cars past when they are in a passing lane, they drive during rush hour and clogg ot up more. Nz the roads consist of a long line of cars behind a truck usually carrying one container and driving alongside an unused rail line. Such a waste destroying the roads, annoying everyone else and making money put of it. The frustrated driver behind then overtakes at the wrong time and causes accident. No blame on slow drivers they are constantly patted on the back and go on their merryway to cause the next accident. Charge truck companies and subsidise rail....hopefully winston deals to them

Having ever larger trucks on our roads is typical of National party ideology. Transport firms having a little more profit, especially if such are not passed on to consumers does nothing for the quality of life of all.

It’s taken a lot of effort to ensure left leaning voters are inconvenienced in every facet of their twisted existence. Thanks for the recognition.

There are many more useful comments than this in the discussion. Suggest you take off your blinkers.

He may have forgotten to add the sarcasm emoji

I’ll have to google how to add the sarcasm emoji. Yes, my comment was intended that way. The OP’s post reminded me of a religious zealot who sees evil everywhere.

Could it be that we're all just being encouraged daily to allow distractions to penetrate our consciousness, that we have to look at our phone whenever it buzzes, that we have to look at all the advertising on the road side, that we pass a driving test by being on our best behavior for 30 minutes then we're good to go, that we have more people in our country than ever before with less emphasis on understanding the laws of physics and more emphasis on a flashy 5 star safety rating with your $40,000 SUV.

Does it matter what the cause is when we know the solution?

Remove the human from the drivers seat.

The quickest way to cut road deaths is the rapid adoption of autonomous vehicles. As a country we have to decide if when we want to make the switch before other countries. And if we do, how do we ensure we get manufacturing priority.

NZ could easily lead the world on autonomous vehicle adoption. We have high fuel costs, highish labor costs, a full range of weather and road conditions, a small but representative population, majority of electricity being renewable. We are the perfect testing ground, but we need government initiatives to court manufacturers, and government investment to make it happen (if we want transport as a service to be publicly owned).

I suspect that by the time computers are truly able to handle all of the possible variables that will arise on our roads they will be able to run our world far better than humans. I fully expect that we will see some truly horrendous accidents caused by AI controlled cars to the degree that their inception will be highly staggered.

How would an autonomous car go on a narrow dirt road at night with hail falling, potholed road, dark coloured cattle running free, a gale blowing the leaves across its sensors path. It might say: data overload! This is dangerous, I refuse to drive. I'm stopping. You walk.

Safely.

really?

Yes really. It'll also be smart enough to drive to the conditions, something our human counterparts forget to do.

They've already worked around rain and snow.

https://qz.com/637509/driverless-cars-have-a-new-way-to-navigate-in-rain...

No unfortunately they have not. visibility in areas with variable roading conditions is still as reliable as a dog left alone with the night's meat for dinner left on the floor. Try reading the research and study conditions and then apply that to most NZ . You can trust it as far as you are there watching it at all times ready to take over at the slightest moment and knowing full well there are several points where conditions and the AI response will lead you into a ditch or, using the analogy, the dog is both faster and more agile than your attempt to stop the theft of meat.

Imagine on this hailing night time trip on the dirt road: just over the brow of a hill is a washout with a raging river down below. How good are these radars? Would they notice the lack of road? Or would it see a 30cm thick trunk of a tree lying on the road?
Or when driving along most dirt roads there are three tyre grooves with piles of dirt inbetween, would the car be able to keep in these and not go onto the marbles? By the way, we all know how useless ABS brakes are at stopping a car on the marbles.
Another one: I have seen a tyre in the middle of a lane on the motorway. Probably a spare from a trailer. I missed it because I wasn't tailgating. Autonomous vehicle: truck to the left, truck to the right, truck tailgating behind: wheel in the middle of the lane: Would this autonomous car slam on the brakes? Or go around it by going very close to the truck on the left like a good driver would?

Actually many self driving vehicles, (note not actually self driving in the giant * they have to list assisted driving), have to drop back to manual control. All those human response mechanisms and inwardly listed rules like give way to cars driving uphill, drive in the center when park cars block the side but pause at an appropriate space to allow oncoming traffic, any condition where the center line or road edge is not well defined, heck you can even trick the AI with some easily repositioned or tagged road signs. In fact all the AI development is not capable outside the strict controlled environment even Tesla's "don't ever take your hands off the wheel and try to second guess when the AI will fail" motto is not going so well. Take the fact that the general populace will give in to believing in the magic of technology and you have a lot of accidents in proportion. Even with fly by wire for pilots and safety of backup mechanisms there are still accidents caused by AI and pilot error recovering from AI failure. Do we trust most humans with computer & system security? No well guess what it will be like when most give up any responsibility and sense of attention to a supposed mystic AI with common developer errors and engineer misconceptions about the breath of case environments. It is really a constant engineering development to reach a completely safe AI system and in planes we have not even made it close so far. But sure around a sleepy American suburbia with easily defined road markings and no major weather issues or introduced risks it may work out ok, (except all those cases of the trolley problems where the death of the driver & or pedestrians is an acceptable result given the choice). It is as safe as a minefield where the map is drawn with a crayon by a monkey, (the developers are represented by the monkey). Heck they still have not figured out kangaroos or black ice conditions. Want to guess what happens when the AI response does not match the environment it is in. Shame you cannot give it a prepare for morons driving in the wrong lane, the road surface giving way or falling rocks mode yet.
The TLDR: https://xkcd.com/1897/

Sounds a bit like our election.

It is just all rhetoric and bluster by those with an adamant position to uphold and political points to score. If police/politicians/roading agencies want to laud their heroic efforts in road toll reduction (as they do) then they should also bear the brickbats for any increase. Otherwise, admit that any efforts are largely pointless and STFU. The large cohort of geniuses who can identify the 1 single cause, be it poor roads, hoons, trucks, drunks, speed, impatience, lack of speed, innattention, fatigue, old drivers, young drivers, foreign drivers, old cars, new cars, gadgets... (what did I miss), you too STFU, it's also pointless, your'e not helping.
Items to note: Cars live longer in NZ because their service life is measured in Km, not years. 1 crash can involve any number of vehicles and potentially kill or injure as many occupants as are in the vehicles. In a single random event you can get a high or low number of victims just like lotto, but no more "crashes" have occurred. There is a very short list of boxes that the police will tick on their report sheet and most of that would be best guess than true fact and actual cause. It is unlikely to be a popular admission that some level of harm must be accepted, but with all things we humans do there is an element of risk. While we should do our best to mitigate what we can, sensationalizing every instance is doing little to improve much.

Thanks Sam for the article. Loss of life in such a way is very sad and the events over the last few days really highlight this and it is completely wrong for us as a society to accept this "toll". Thoughts and sympathy to all families and friends affected by this.
Assuming everyone is a good driver, we're all human and sometimes we're not so good or maybe you're perfect and someone coming the other way is having a bad day, they are tired, distracted, it doesn't matter but it could happen to any of us. Of course driver training, enforcement, better cars all help but people still make mistakes. Research has shown that is everyone obeyed the road rules we would still have a substantial road toll. Around 4O% of fatal or serious crashes on rural roads involve vehicles running off the road.
In a crash into a tree, power pole, cliffside at 100km/h 8 out of 10 would result in either fatal or serious injury at 80km/h this almost half's. Until we have autonomous vehicles, there are really only two choices we either upgrade all rural roads to motorway standards (allows risk to reduce significantly whilst maintaining speed) or reduce speed with respect to the standard of the road ( not popular but it works ).
Really the only answer we can accept is zero for our road toll.

Sam,
http://www.transport.govt.nz/assets/Uploads/Our-Work/Documents/Summary-o...
page 37
114. For example, in 2016, provisional road toll data indicates that:
• 42 percent of fatal crashes had drugs and/or alcohol as a contributing factor
• 39 percent of drivers and 42 percent of passengers killed in car accidents were not wearing seatbelts

Also if I remember correctly about 94% of all crashes come down to human factors, about 4% the state of the vehicle and about 2% other factors like the road.

That said there is a lot NZ can do to road design.
The Swede's fatality rate is about 1/3 NZ (2.0 per 100,000pop vs 6.0 https://en.wikipedia.org/wiki/List_of_countries_by_traffic-related_death...) and they have a vision http://www.visionzeroinitiative.com/

NZ doesnt even have a separate activity class for funding for safety works (we do for safety education). Its simply absurd.

If we bothered to put congestion tolling in place in Auckland (at a minimum) we could delay & spread out over time the hugely expensive urban highway projects & spend it on safety instead.

"Also if I remember correctly about 94% of all crashes come down to human factors, about 4% the state of the vehicle and about 2% other factors like the road."

94% of accidents might start with someone making a mistake, but roads could be safer once those mistakes happen.

This study (http://www.transport.govt.nz/assets/Uploads/Research/Documents/Long-Term...) previously attributed the downward trend in the following way:
• 45% are attributed to the vehicle: improvements in vehicle crash worthiness and fewer motorcycles (which are in effect less safe vehicles) per unit of travel.
• 19% are attributed to the road, measured as real net investment in roading per unit of travel
• 36% are attributed to the driver; a combination of the effects of advertising, breath testing, and average speed.

If you want to decrease the road toll, change the side of the road we drive on so that everyone has to actually concentrate while driving. :-)

https://en.wikipedia.org/wiki/Dagen_H#Results