sign uplog in
Want to go ad-free? Find out how, here.

Sam Warburton correlates the reduced numbers of Police on the roads with the rising road toll

Sam Warburton correlates the reduced numbers of Police on the roads with the rising road toll

By Sam Warburton*

The last time I wrote for Interest about the road toll was in October. At the time, Police were predicting 60 further deaths for 2017. The year ended with 86 further deaths and a final toll of 379 lives – about 10% higher than 2016 and 40% higher than 2013 even after accounting for more driving.

The Police were in the news again this week thanks to an investigation by Radio NZ showing that road Police numbers had fallen even as the risk of dying has increased.

Police Association President Chris Cahill said ‘Without a doubt, there has to be some correlation… The limited or lesser amount of road policing enforcement being done will be having an effect on the road toll’.

Former Assistant Policy Commissioner Dave Cliff told Radio NZ that this was a ‘serious concern… [because] the deterrence value of a police officer, the deterrence value of speed cameras, is major’.

Cahill’s comment will be true – any spending on Police will have some effect on the road toll – but how much depends on the accuracy of Cliff’s comment that Police are a major deterrent.

As National’s Police spokesperson Chris Bishop pointed out in the same article, Police are but one of many factors that influence the risk of death or injury.

Let’s put some of this in context.

Expenditure on road police

Every three years, the Government issues a Government Policy Statement setting out its objectives for transport and providing funding signals to the NZ Transport Agency and councils. This includes funding for road policing which comes from the National Land Transport Fund. What the Police is contracted to provide is negotiated between Police and the NZ Transport Agency.

In 2012, the Government issued signals that suggested it would be happy with a small reduction in road police spending. The NZ Transport Agency acted on those signals and, after accounting for inflation, provided 2.4% less funding to Police.

This was reversed in 2015 and funding increased by 4.7%.

Overall, inflation-adjusted funding for road policing has been pretty flat between 2009/10 and 2017/18.

The impact on Police effectiveness will be large

On the face of it, flat expenditure and a gradual decline in the number of Police by 7.6% over 10 years might not seem like a big deal.

For comparison, the amount we travelled as a country reduced slightly between 2009 and 2012 and the incidence of death per kilometre driven continued its historical decline. A small reduction in policing might not have made much difference.

Further, if we’ve swapped officers for lower-cost speed cameras, it’s possible that the level of road policing hasn’t materially changed and allowed those officers to be redeployed to other Police activities.

However, we’re also driving much more now. Kilometres travelled in 2017 are expected to be around 15% higher than in 2013. And every kilometre we drive carries a 40% higher risk of death than it did in 2013.

Police are also funded for a whole lot of things speed cameras can’t do. Police issue tickets for drink driving, cellphone use, not wearing seatbelts and other high-risk behaviours. They conduct drink driving tests, assist at accidents and give road safety sessions at schools.

The table below is representative of the impact on Police outputs:

  2016/17 2017/18
Output Target Actual Target
Enforcement (e.g. tickets, prosecutions) undertaken for:
'High-risk' driving offences per 10,000 population 775 to 825 790 450 to 600
Speed per 10,000 population 590 to 635 630 555 to 600
Not wearing seatbelts per 10,000 population 100 to 125 120 94 to 120
Mobile-phones per 10,000 population 40 to 60 60 38 to 56
Crash attendance and reporting
Number of emergency traffic events responded to 28,000 to 38,000 48,000 40,000 to 50,000

In 2016/17, Police had performance targets such as issuing about 800 tickets for ‘high-risk’ behaviours per 10,000 population. For 2017/18, that target has dropped to about 500 tickets. Almost every performance target (full set here) has dropped.

This would be acceptable if Police enforcement was achieving its goal of reducing risky behaviour, but we’ve had months of Police imploring people to drive more safely while the incidence of death continues to rise. Even seatbelts and cell phones, two factors widely cited as needing addressing, have seen reduced enforcement.

Actual enforcement actions tend to be at the top of target ranges. The maximum of the 2016/17 target was reached in the case of cell phones. This suggests that the target prevented Police from doing more to address cell phone use. In 2017/18, the target has reduced.

One of the few actions Police have been funded to do more of is attending emergency traffic incidents – from about 33,000 to about 45,000.

It seems that enforcement has been reduced so that Police can attend a growing number of accidents.

If Police enforcement makes a reasonable contribution to safety, this doesn’t bode well for 2017/18’s road toll.

The impact on overall policy effectiveness is uncertain

Road policing wasn’t a special target for cost-saving in 2012. Essentially every activity, from public transport to running the NZ Transport Agency, was reduced to free up money for state highway improvements, dominated at the time by the Roads of National Significance.

Figure: Funding ranges by activity class, 2009/10 to 2017/18

Funding reductions also occurred for road safety promotion (e.g. advertising, education) and, likely, state highway safety improvements. Last year I noted that improvements to dangerous stretches of roads had been delayed because funding was directed elsewhere.

The reduced focus on infrastructure improvements, advertising and education, at the same time as policing, makes it difficult to draw conclusions as strong as the Police Association President might like.

But it shouldn’t be so difficult for the Ministry of Transport and the NZ Transport Agency. These agencies are resourced to monitor and evaluate safety interventions and advise on such things as whether Police should be a funding priority.

Why the secrecy?

Recent media coverage of the road toll has seen dozens of interviews with government employees. Some have expressed exasperation. Some have pointed to studies they’ve commissioned that conclude the road toll’s still going down and there’s no problem. Some have speculated as to causes, but only those within drivers’ control.

No agency – not the Ministry of Transport, the NZ Transport Agency or Police – talked about the reduced expenditure on road safety promotion, road maintenance and improvements. Even if they did, none would have been able to say which interventions are most cost-effective.

Certainly, no agency mentioned declining Police numbers. Despite huge public interest, it took Police three months and two Official Information Act extensions to release the data to Radio NZ.

The Ministry of Transport’s Briefing to the Incoming Minister didn’t mention the issue at all. If the Policy’s briefing did, it has been kept it from the public:

Agencies are required to provide free and frank advice to Government and to be transparent with the public. Agencies are also meant to aide public discussions about issues like these and guide people through the choices we have.

The next big chance to do that will be in consulting on a Government Policy Statement early this year. Agencies are going to have to lift the quality of their advice and be a lot more transparent if we’re to get the best from our transport system.

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


We welcome your comments below. If you are not already registered, please register to 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.


One of the things that frustrate me with this topic of discussion is the lack of detail. There is so much missing detail around the fatalities such as where are they occurring, what speeds are involved, what are the demographics of the drivers, what type of vehicles and so on. we have all seen Police monitoring traffic on straight roads which last saw an accident 10 or so years ago, or not at all. Just throwing money at a problem won't necessarily solve it. Having a strategy around the real factors, in depth, should have an impact.

But also at the end of the day neither Police nor legislation cannot prevent STUPID from getting into the car, and there is plenty of that happening every day! Perhaps that is a topic that could be discussed in schools?

It is no different to policing a city properly
The NY police immediately target a area where there is crime and obliterate the main cause in that area
Traffic should be dealt with by using statistics in the same manner
There’s also a government directive for police to collect revenue which distorts where traffic police should be policing. Revenue collection opportunities are high on the agenda sharing time with more important targets that could lower fatalities & accidents.
I personally would not place revenue collectors on a busy morning peak hour intersection yet that’s what happens in Auckland I’ve experienced myself.
There needs to be a demographic & age & sex breakdown along with isolation of worst accident points in the reading system. It can be solved but not if priorities are distorted by government demands for revenue collection albeit government may protest otherwise.

Lots of reasons for the high road toll.
The ones that I think are prominent are rumble strips on the left which force vehicles to be much closer than necessary to oncoming traffic.
General frustrations and reduced overtaking options such as all the speed advisory signs which some people take to be the gospel even on bright sunny days, converting overtaking lanes to "slow vehicle lanes" (eg desert road) where very few vehicles ever venture, increasing the frustration levels of following traffic.
And this recent fetish of buying vehicles with very high centres of gravity etc etc

... yes , those rumble strips on the left are sheer crazy ... they ought to be up the centre , dividing the lanes ... encouraging opposite direction traffic to pull away from each other ...

Also ... there's many seat-belts around the country which must be way way too tight ... 'cos as soon as the driver gets into their car , the belt constricts bloodflow to the brain ... the brain ceases to function ... and the driver begins their journey .... VROOOOM VROOOOOOOOOOOOOOM ....

The centre that is the biggest problem on our roads is self centred. Add impatience to that, and you have a potent mix.
I would, however, agree with you about vehicles with high centres of gravity and I would add to that the proliferation of 7 and more seaters, often with all seats occupied, thus increasing the number of people involved in many accidents.
Oh and of course, drunk/drugged driving in the wee small hours and tourist drivers in certain parts of the country.

Well its to be expected . After all we are registering thousands of additional cars on the roads each year , and the volumes of traffic have increased in tandem with this .

This is not to say we should not be outraged by the death toll , and maybe its time to increase the number of fixed speed cameras at known accident hotspots as a first step .

They don't put them there because the vehicle count is not high enough to make them economical.......

That was the original sales pitch for speed cameras, from what I recall. That they would be placed at black spots to address this issue.

Unfortunately it seemed later that a lot of these black spots were mysteriously long straights or slight downhills where speed naturally increases somewhat and people have never seen an accident in two lifetimes.

On the camera note, though, it's time we had more red light cameras in central Auckland now. Things have gotten absolutely ridiculous. Red lights are seen as completely optional.

It can't be a linear relationship. As a dumb example let's say there are 10 cars passing 10 other cars on a straight road, that's 100 conflicts. Let's say the population grows 20%. 12 cars passing another 12 cars = 144 conflicts and each car is now following closer to the next car. A barrier might fix this on a highway, but what about level intersections?

It is a closed shop for a reason. I suspect detailed analysis would show up their waste of resources on a wide range of activities. If they claim success in reducing harm through their efforts they must also accept blame for an increase. They have beaten to death the few available tickable boxes and a glance outside the box is simply not permitted. The latest campaign is more tiresome speculative and emotive scaremongering which will only serve to cow the feeble minded and justify the actions of the timid whose inept blockading frustrates the competent. The sooner road policing is automated and we are spared the patronizing lectures in lieu of actual effective strategy the better.
Here are some other high quality stats, something simply must be done!!!!!!!!!!!!!!

Wouldn't it make more sense that if you have more vehicles per road (be it length or number of intersections) there will be more crashes? The relationship of population to vehicle interactions is not linear especially when people are forced to buy in the suburbs and have longer commutes. Roads can only scale so far until you need expensive barriers and flyovers. National decided to cram in more people but not put in the very costly infrastructure.

I would like to see population compared with deaths. This study seems to agree that more traffic = higher crash rate:

In many areas of government and policy, there are facts being concealed, and one that is being concealed in this policy area, is that it has been proven again and again that the safest speed limit is "the 85th percentile speed at which everyone is driving at their own discretion". Set the speed limit higher than that, and it is less safe; set it lower than that, and likewise it is less safe.

I hold the authorities responsible, on this subject, for creating a whole new class of inconsiderate drivers - a few decades ago, I remember that "dawdling" was universally regarded as inconsiderate, and bad practice. Now dawdling is a form of virtue-signalling. It does not seem to matter that most of the people who do it are dangerously incompetent drivers, period: there are no legal penalties to being dangerously incompetent.

One of the reasons Germany (and some other European countries) have such high speeds and such low crash rates, is that the standard for getting a drivers licence is so high (and expensive and time-consuming) that a significant minority simply cannot even get one - competence, co-ordination, concentration, etc matter very much at the standard applied (eg test-track high-speed driving, skid control, etc etc). This is also a reason why so many people over there use public transport or bicycles - the people who can't get a driver's licence are a significant captive customer base for non-driving "transport" modes.

In Wellington the camera cars and hand held police speed cameras seem to be mostly used on 4 lane motorways with a barrier. The danger of speeding on these roads seems to be very low - one year (perhaps 2015?) there were only 2 deaths on these 4 lane median barrier roads out of 300 plus fatalities. I think the idea is that is you speed on a motorway then you will speed everywhere - based on some report about 20 years ago. But all it seems to do is lose the support of the public and then you get your speed fix in a more dangerous road. Also for a lot of people the fine is worse than the crime - so the overall damage to the public is greater than any safety benefit of slightly slower cars on the motorway.
You just dont get the feeling the police or bureaucrats or politicians have the slightest clue in what they are doing.

In Wellington the camera cars and hand held police speed cameras seem to be mostly used on 4 lane motorways with a barrier. The danger of speeding on these roads seems to be very low - one year (perhaps 2015?) there were only 2 deaths on these 4 lane median barrier roads out of 300 plus fatalities. I think the idea is that is you speed on a motorway then you will speed everywhere - based on some report about 20 years ago. But all it seems to do is lose the support of the public and then you get your speed fix in a more dangerous road. Also for a lot of people the fine is worse than the crime - so the overall damage to the public is greater than any safety benefit of slightly slower cars on the motorway.
You just dont get the feeling the police or bureaucrats or politicians have the slightest clue in what they are doing.

Driving down Rifle Range road in Hamilton the other day i came across a policeman holding a hand held radar.
Meanwhile 100 yards behind him is a roundabout which sees near misses and also crashes on a regular basis.

1) Speeding - it would be a better use of resources to implement an extensive network of speed and travel time cameras. Manual enforcement is old technology and inefficient. Police time would be better used for other more important matters like drink driving given the still high number of drink related crashes.

2) Seatbelts - there are speed cameras where I reside which can capture seatbelt use. So same as 1). Lack of seatbelt use is also a large cause of road fatalities and injuries

3) Use of cameras would allow an sms to be sent immediately (or delayed) to the vehicle owners phone or email.

4) Ban vehicle reregistration until the fines are paid as they do where I reside.