Knock, knock, knock. It's the police, open the door.
What do you want?
We just want to talk.
How many of you are there?
There are two of us.
Then why don't you talk to each other?

Even when they're not knocking on your door, we should be concerned by police numbers.

In the thick of an election campaign, there will be some wild policies bandied around with regard to crime and justice - don't get me started on boot camps - and a sober debate around police numbers is hardly likely to set the electorate alight. But it ought. That's if we are to attempt a radical rethinking of what policing means - and I believe we should.

After significant growth between 2001 and 2008, police numbers have plateaued around 8800 since 2008. Population growth has seen the ratio of police to population shrink from 1 in 500 to 1 in 530. Crime resolution rates have dropped recently too.


Both National and Labour have responded by promising large increases in sworn police numbers: by 880 and 1000 respectively.

The rise in numbers is welcome, but such a dramatic increase will be difficult to achieve within the political timeframes. The capacity to recruit the right people (filling the various representation targets) and then train them without dropping standards makes that job nigh on impossible. The drive, however, is important.

The police have set themselves some bold targets, not the least of which is to reduce Maori offending by 25 per cent by 2025. New results, however, will not stem from old methods.

Police are currently moving toward a community-policing model, and if done correctly it could be a policing revolution. The old school cop's only goal was to get as many bad guys in cuffs and in the cells as possible. The new focus, in part, is about crime prevention. And this means looking at and addressing crime in more sophisticated ways.

A simple shoplifting case, for example, may be as much about school truancy as theft, family violence may be as much about addiction, and all manner of antisocial behaviour may be about mental health. A remarkable 20 per cent of calls to the police involve mental health issues. Making arrests in these instances may solve an immediate problem, but the police will be back time-and-time again to arrest the same person if the underlying issues remain unaddressed. As the Police Association has made clear, "we cannot arrest or imprison our way to safer communities". Police therefore need to work in with the likes of the Ministry of Education, the Ministry of Health and others to identify and then correct deeper family and community concerns.

Increasing police numbers will make such an approach viable.

Furthermore, outside of specific investigations, it's desirable to see the police about. In our malls chatting with people, in schools letting kids play with handcuffs, the community needs to know their police and the police need to know the community. Police work best when they police by consent.

I'm not always going to agree with all of the police's activities or actions. I've called them on a number of issues and will continue to do so (incidentally, I'd appreciate it if they changed the fact I'm recorded as being a Hell's Angel in their computer system), but overall police can have an important role to play in our society working toward broader goals of making the community a better place and, critically, reducing our burgeoning and expensive prison population.

Many people will say that this is not the role of the police, and that they should stick to their knitting. But if we continue to do the same things, we will get the same results. We need to be bold and innovative.

Crime is undoubtedly a complex phenomenon. We are not going to solve it with simple solutions. We ought be suspicious of politicians peddling simple "common-sense" solutions that have been tried and failed before. Any credible analysis of crime will show that only a multi-agency approach and wrap around services of at-risk families and within at-risk communities will have any credible chance of solving problems that have often taken generations to create. In my view the police not only have a role here but they have an important role. But that means fundamentally rethinking what policing is, and ensuring that police have the numbers to undertake it.

Dr Jarrod Gilbert is a sociologist at the University of Canterbury and the lead researcher at Independent Research Solutions. He is an award-winning writer who specialises in research with practical applications.