Skip to Content »

Spending on crime vs spending on education

  • March 12th, 2008

Thomas Sowell’s latest column goes crisply to the heart of a common fallacy, that crime and education spending are substitutable.

I wish he’d sourced his figures, but they are in line with the only comprehensive New Zealand study of the costs of crime. NZIER’s draft report can be found, but the government stopped it there. It would have told us that crime costs (control, precautions and victim losses) amounted in the 1990s to over 5% of GDP. It would be around $8 billion today.

[Update – That is plainly a gross under-estimate. A Treasury working paper put the 2003-2004 cost at $9.1 billion of which $7 billion was private cost and loss. Thanks Eric Crampton (whose comment below alerted me to this). I wish I’d come across that paper when I was an MP ]

If only prison and educaton costs were substitutable.  19th and 20th century progressives were firmly convinced that once wealth was spread properly, so that even the poor could read and write, the criminal classes would be educated out of crime. It seemed a reasonable expectation. Sadly, the facts have mugged that hope, just as they stabbed the belief that crime would wither when everyone had a home and enough to eat and wear. The latter took a beating when depression and recession statistics showed reductions, not increases in crime. Still, many progressives believe in their hearts that violent predation is an understandable class reaction to poverty, and we have just not been nice enough for long enough to criminals to test their theories properly.

Sowell mentions David Fraser, who will visit New Zealand this year as a guest of the Sensible Sentencing Trust. Fraser has infuriated the self anointed criminal justice elite by lookng at the actual results of their goofy theories, rather than their fond hopes. Worst of all he insists on ignoring the other great progressive fallacy,  that penal policy should be measured by its success in rehabilitation.

The only valid measure is in fact change in crime and victimisation rates.

The international evidence is overwhelming. Rehabilitation happens, but for adults it is almost always spontaneous (no positive effect of programmes and release policies). Rehabilitation is a seductive but distracting primary goal for penal policy. Sadly it has become the only goal of New Zealand’s justice anointed. In the face of failure they redouble their effort, and the shrillness of their denunciation of doubting unbelievers.

While they fail victims suffer our spectacular increases in violent crime rates.


  • mollygog
  • March 12th, 2008
  • 9:52 am

nicely said
i agree with it all and fail to understand why the government doesn’t to?

while crime and education spending aren’t substitutable I do believe that if a child leaves school illiterate you almost consign them to a life of crime.

If you can’t read or write your choices are extremely limited. You can work as a labourer, choose not to work at all, or work as a criminal. You wouldn’t even need to do a proper study – i think we all know there are huge levels of illiteracy in the unskilled workforce, beneficiary and prison populations.

So why doesn’t government screen for learning disabilities and other factors which prevent learning? Why don’t they test for levels of literacy? And then throw in a bit of extra money at that to ensure as few students as possible leave school without basic reading and writing skills?

hope to see you back in parliament soon


Stephen, check Treasury’s estimate available here:

They reckon costs in 2003/4 to be $9.1 billion, $7 billion of which falls on the private sector. They also tell us that a 1% increase in crime volume results in an increase in costs of $74 million.

Now, some fun back of the envelope stuff. If the Klick and Tabarrok US estimates of elasticity of crime with respect to police presence can hold here in NZ, it would take a 3.33% increase in police numbers to reduce crime by 1%. I’d think that NZ would be more responsive to changes in the number of police than the US numbers would indicate, as we have a lower policing rate than does the US and there’s diminishing returns, but let’s stick with the US numbers as a lower bound estimate of effectiveness. Would it cost $74 million to increase the number of police by 3.33%? The Treasury stats put total police costs at $872 million. A very rough estimate would say that if we increase the police budget by 3.33%, we should be able to get 3.33% more officers. That would cost $29 million. Add in the deadweight costs of taxation: it costs us more than $29 million to get $29 million in taxes. Once you add in the deadweight losses of taxation, it looks like it would cost us $38 million to get a crime reduction of $74 million. Clearly, this is very rough. But as a first cut, it looks like a dollar’s extra spending on police saves the country $2 in crime. Policy conclusion: hire more police officers.

  • Stuart
  • March 13th, 2008
  • 9:29 pm

Stephen, if you’re an MP later in the year (and I hope you are) do you believe you can effect real change in this area?

As you know, the entrenched and bureaucratic justice/corrections system desperately needs an overhaul. Cheers Stuart


Newbolt sums up the last failed experiment [see below]. I hope Judith Collins doesn’t fall for the same con job.

1. Newbold, Greg. “Integrated Offender Management: New Zealand’s Latest Experiment in Criminal Rehabilitation” Paper presented at the annual meeting of the AMERICAN SOCIETY OF CRIMINOLOGY, Atlanta Marriott Marquis, Atlanta, Georgia
Publication Type: Conference Paper/Unpublished Manuscript
Abstract: Since 1910, New Zealand has been engaged in a constant search to find a method of rehabilitating criminals that really works. In 1996, inspired by the work of Canadian criminologist Paul Gendreau and others, the Department of Corrections embarked on a new experiment called Integrated Offender Management (IOM). Based on a psychotherapeutic model, IOM involves a complicated and expensive process of identifying an inmate’s ‘criminogenic needs’, creating programs to address those ‘needs’, and applying the programs in the hope of preventing further offending. When initially conceived it was hoped that IOM would produce at least a 25 percent improvement in overall correctional efficiency. Eleven years on, with five-year reconviction rates remaining in the region of 86 percent, it appears that IOM has failed. This paper examines the objectives, strategy, and actual implementation of IOM in New Zealand, and suggests why the project inevitably foundered.

Leave your comments:

* Required fields. Your e-mail address will not be published on this site

You can use the following HTML tags:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>