The real question for policymakers is not whether prison works, but whether it is the best possible way to spend scarce resources.
Lets take a healthcare analogy. Suppose we live in a society that tolerates abysmal sanitation, and because of this, suffers predictably high levels of preventable disease. Under these conditions we will "need" to build a lot of hospitals to care for the sick and keep them away from others – which the hospitals will to some extent do, and therefore "work." Yet people will still be getting sick from a lack of prevention.
So it is with violence. If we accept the conditions that predictably breed large amounts of violent crime – record levels of inequality, for example, or broken mental health systems – then we will "need" a lot of prisons. Building more may even prevent some crime. But the approach will still be irrational because there are far better alternatives.
New Zealand is in the middle of a housing crisis. We have built six prisons since 2005, some the size of small towns. On current forecasts, we will need another every three years. The Government is scrambling for resources and construction capacity to build affordable homes. I hope they will ignore the tired get tough slogans from National and Garth McVicar. Building houses will work far better, making our people safe, than building prisons.
We should start with an honest assessment of the evidence. From there, policy choices become a question of values about the kind of society we want New Zealand to be. Houses not cages wins on that count too.
• Liam Martin, criminologist at the University of Victoria.