There have been no third strikes, yet.

Offending by young people, and even those recognised by the criminal justice system as adults at 17, does not fall within the three strike regime.

People between the ages of 18 and 24 are young people as far as I am concerned. Certainly as far as cognitive development is concerned. So I don't think being pedantic about the definition of "young person" is helpful.

But that does raise quite an important point. Do your numbers account for the fact that people committing strike offences in 2005-2010 include young people under 18 years old? Again - possibly very significant impact on the results.

I don't think that it would be odd, based on what we understand about the drivers of recidivism. Young people can commit first strike offences out of sheer stupidity and impulsiveness. When they are sentenced, there will be various risk factors that indicate a tendency to re-offend. If there has been intervention early-on those risk factors are less likely to be evident: recidivism will go down.

In any event, the numbers show that there has been a reduction in strike one offending consistent with demographic trends. Ultimately you would expect recidivism to fall too and there are various reasons it might fall faster than the overall strike rate.

Thank you for the interesting post. As you rightly foreshadowed, people sceptical of 3 strikes (like myself) are unlikely to be convinced that you have demonstrated a specific deterrent effect though.

As always, there are a number of explanations.

One I thought I would throw out there: the combination of factors that lead to it being less likely that someone will be charged with a second strike offence. The Criminal Procedure Act 2011 and the bulk funding of Crown prosecutors (since 2008?) have both increased the focus on 'plea bargaining'. Where an accused faces a second strike offence, plea bargaining becomes obvious. Defence and Crown will negotiate the charge down to a non-strike offence and a guilty plea. Eg Agg Robbery/Burglary is downgraded to standard Robbery/Burglary. This happens a lot, the incentives are evident for the Crown and the defence. Most of the lawyers involved don't like 3 strikes anyway.

That factor alone probably does not account for the effect you demonstrated, but I suggest that it is quite significant.

Finally, I think it is wrongheaded to look for the cause of reduced recidivism in programmes implemented recently. Much more likely it would be programmes implemented with children about 10-18 years before they end up in prison for the first time. That is, I would look for intervention in 1997-2005. Only secondarily would I look at the services provided by Corrections.

And so it goes on...