One Question Quiz
Photo: Montage of NZ Police image
Photo: Montage of NZ Police image

SocietyMarch 30, 2017

Piling cash into boosting police numbers is pointless, and this graph proves it

Photo: Montage of NZ Police image
Photo: Montage of NZ Police image

The evidence shows that a ‘tough on crime’ approach is a posture, not a solution, writes criminologist Antje Deckert.

In 2011, when Bill English was Finance Minister, he declared that New Zealand’s prisons were “a moral and fiscal failure”. Five years later, the National government has announced that it will recruit 1,100 new police officers and spend $1 billion on 1,800 new prison beds. Having stepped up to the role of prime minister, Bill English now exudes a “tough on crime” persona, apparently turning a blind eye to the devastating fiscal consequences of a law and order agenda.

What does it cost the taxpayer to add 1,100 sworn officers to the police force?

The starting salary of a sworn police officer straight after training is roughly around $61,500 per year, which includes benefits and allowances. Hence, the 1,100 new officers will cost the taxpayer an additional $67.7 million each year. This is, however, only the salary cost and excludes funds that will have to be spent on their training, uniforms, equipment and police cars.

What can the taxpayer expect for that amount of money?

In a nutshell: Not much. The graph below demonstrates why. The blue line represents the number of police officers for every 10,000 New Zealanders. To maintain a balanced ratio of civilians and police officers, governments tend to employ more police officers when the population grows and put a hold on new officer recruitment when the population declines. Between 2005 and 2014, the number of police officers per 10,000 New Zealanders remained fairly consisted with 18 to 20 officers for every 10,000 New Zealanders. The announced increase will bring this ratio to an unprecedented 22 per 10,000 inhabitants.

The black dotted line shows the number of recorded crimes per police officer and the grey dotted line shows the number of arrests made per police officer. Both run fairly parallel, which makes sense because fewer people should be arrested when fewer crimes are recorded. In other words, if the number of reported crimes goes up, each police officer gets busier. If the number of reported crimes goes down, police officers can dedicate more time to each case.

The green line shows how many of 100 recorded crimes were resolved by police. A case is marked as resolved when police have identified a suspect. Of 100 cases, police tend to solve between 42 and 47. It seems rational to assume that increasing the number of police officers would result in more crimes being solved. This is, however, not the case  – as a look at the green and blue line shows. In 2005, we had 7,385 sworn police officers (18 per 10,000 inhabitants) who identified a suspect in 44 of 100 cases. In 2013, we had 8,702 officers (20 per 10,000 inhabitants) who also found a suspect in 44 of 100 cases. So, although we had 1,317 more officers on the street in 2013, no more crimes got solved than in 2005. Hence, we must assume that raising the number of police officers by another 1,100 (22 officers per 10,000 inhabitants) will achieve just as little as the increase in police staff between 2005 and 2013.

The main question then becomes: why does hiring more police officers not result in more crime being solved? Here are three key reasons: (all crime statistics from

1) All the crime types that are easier to solve have declined in numbers

“Easier to solve” means that these crimes have a high resolution rate, ie the police can identify a suspect. Largely, the resolution rate depends on the circumstances that commonly surround a specific crime type. Murder, for example, has a resolution rate of nearly 100 percent because a dead body or missing person is usually reported immediately. Often there are witnesses to the murder or next of kin can name a suspect with motive. The number of murders has been in steady decline. In 2005, police recorded 61 murders; in 2014 the number was 41.

2) Nearly all crime types that are harder to solve have also declined in numbers

“Harder to solve” means that in most reported cases no suspect can be named. Shoplifting, for example, is often detected after the fact. Rarely are shoplifters caught red-handed. In most cases, shop owners realise that something went missing when they do their regular stocktake. Therefore, the resolution rate hovers around 30 percent each year. Of crime types, all but two (see below) have also declined over the past decade.

3) The only two crime types that have increased over the past decade are ‘harder to solve’ crimes

The only two crime types that have reportedly increased are domestic violence and sexual assault. Since these two offence types typically occur behind closed doors, police will be unable to prevent them by showing more presence on the street. They will also not be able to solve more cases after they have occurred. The resolution rate for sex crimes is represented by the orange line in the graph. On average, police identify a suspect in 48 to 64 of 100 reported sex crimes. Because most sexual offences happen behind closed doors, police rarely have a witness to rely on.

Often the victim’s word stands against the word of the alleged offender. If there is no witness or physical evidence, police frequently do not have enough to identify the alleged offender as the suspect. To charge someone with an offence, police must gather enough evidence to believe that the person faces realistic chances of a conviction. Obviously, the circumstances of sexual crimes will not change by hiring more police. Hiring more police will also not get rid of the evidence problems that arise from these circumstances. This is confirmed when looking at the orange and blue line in the graph. They highlight that the percentage of resolved sexual offences has declined, while the number of police officers per 10,000 New Zealanders has increased.

Unlike a judge or jury, police are allowed to have doubts about someone’s guilt. This is reflected by the purple dashed line in the graph. On average, courts dispense 40 to 44 convictions for every 100 arrests police make.

Interestingly, the graph shows that the number of crimes (black line), arrests (grey line) and identified suspects (green line) has declined, but, nevertheless, the number of convictions per arrest (purple line) and the number of convictions per court cases (red line) have increased. In other words, we have arrived at a paradoxical situation where fewer crimes are reported but more people are convicted. In 2005, courts convicted in 69 of 100 cases. By 2014, that number had risen to 76 of 100 cases. Also, the use of custodial sentences (imprisonment, preventative and home detention) has increased from 11 to 15 in 100 convictions. In other words, 6 percent of the New Zealanders we lock up today would have received a community sentence, diversion or similar response in 2005.

Evidently, New Zealand has grown more punitive over the last decade. Adding 1,100 more police officers to the streets only serves to support this ‘tough on crime’ image. It may make voters who are already safe feel safe, but it does absolutely nothing to protect actual victims of domestic and sexual violence, the two crime types whose numbers continue to rise.

Investment according to current crime trends

It is very unlikely that an additional 1,100 police officers will prevent or solve more crime. Overall, crime has been decreasing in the Western world – not because we have more police on the streets but because we have fewer people in the population that are in the high-offending age group (i.e. adolescents and young adults). This downward trend is going to continue regardless of how many police officers patrol our streets. Yet New Zealand taxpayers will be burdened with over $67.7 million in additional expenses each year.

Instead of hitting a screw with a hammer, we should look for the right screwdriver in the toolbox. Instead of blindly increasing the number of police officers, we should be investing that $67.7 million according to current crime type trends. In other words, we should make a targeted investment when it is required and where our investment guarantees a return on our investment. Hiring more police officers would be a good investment if the murder rate was going up, because murder has a resolution rate of nearly 100 percent. But the murder rate in New Zealand has decreased significantly and the two crime types that continue to increase cannot be solved or prevent with more police power.

The annual economic cost of family violence is estimated to be $8 billion. As a nation, we cannot afford to spend taxpayer money on prevention strategies that we know are not going to work. In the current situation, the $67.7 million per year would be better spent on programmes that are proven to help prevent domestic and sexual violence. In this investment scenario, the initial costs would eventually be offset because as fewer of these crimes are committed, we would spend less on medical and psychological treatment for victims. It would also lower expenditure on police, courts, and prisons because prevented domestic and sexual violence means fewer call outs, arrests, convictions and prisoners. And it would lower costs spent on murder investigations – because 40 percent of murder victims are repeat victims of domestic violence.

New Zealand taxpayers have worked hard for their money and have therefore a vested interest in politicians spending tax funds wisely on strategies that work. To borrow author David Rothenberg’s phrase, “being tough on crime is posture, not a solution”. Invest in solutions, New Zealand.

Dr Antje Deckert is a senior lecturer in criminology at AUT.

The Society section is sponsored by AUT. As a contemporary university we’re focused on providing exceptional learning experiences, developing impactful research and forging strong industry partnerships. Start your university journey with us today.

Keep going!