Crime went up last year, the biggest real increase in years. Murders were up by 25 per cent, the 65 recorded homicides being the highest in a decade. And violent crime was up by 8 per cent per capita.

Why? Police put much of the growth down to changes in reporting trends, particularly in family violence, but that doesn't tell us much about the driving forces.

Is it too simplistic or criminal-excusing namby-pambyism to point to the corresponding rise in joblessness last year?

Apparently - though not too simplistic to expect the National-Act "three strikes" law will deal a decisive blow against violent crime.

The Sensible Sentencing Trust's Garth McVicar, that plain-speaking, common-sensical hero of crime victims, promises the streets of New Zealand will be much safer once the Sentencing and Parole Reform bill is passed.

So does his brother in arms, Act MP David Garrett, an ardent admirer of America's three strikes laws which he claims is responsible for falling crime rates in the US in the 1990s.

Garrett wants us to be like California, where a particularly draconian version of the three strikes law has seen a 50 per cent drop in convictions for homicide and robbery.

That may be a little difficult, considering our version of three strikes is different - thankfully, considering the non-parole sentences handed out there, of 50 years for stealing $150 worth of children's videos, and 25 years for stealing golf clubs.

And considering, too, that the reasons for the remarkable drop in crime rates in California and indeed throughout the US since the early 1990s are highly disputed.

Many reasons have been advanced for the dramatic fall in US crime rates.

Probably the most controversial - and hotly contested - was the suggestion by American economists Steve Levitt (of Freakonomics fame) and John Donohue that abortion accounted for as much as half the decline in crime rates.

They contended that abortion led to fewer unwanted babies, who were more likely to suffer abuse and neglect and were therefore more likely to end up as criminals.

In a 2001 paper, they argued it was the absence of unwanted aborted children since abortion was legalised in 1973 that had led to a reduction in crime 18 years later, starting in 1992, when the evidence showed that crime in the US had started to decline. (Males aged 18 to 24 being most likely to commit crimes.)

This prompted the widely-condemned statement by a former US secretary of education that "if you wanted to reduce crime - if that were your sole purpose, you could abort every black baby in this country, and your crime rate would go down".

Which isn't too far a cry from the sterilisation option mooted by David Garrett and others in this country.

Much has been made, too, of the crime-cutting role of tough policing under zero tolerance and "broken windows" policies adopted by the likes of New York mayor Rudi Giuliani. But a number of scientific studies have since showed this was much less significant than first thought.

Disputed, too, is the impact of the massive increase in prisoners that followed the introduction of three strikes laws in 22 states.

Critics point out the decline in US crime started before the introduction of three strikes (1994 in California ) and that states like New York, which never introduced three strikes, had much sharper declines than California.

In other words, as The Sentencing Project pointed out in a 2005 paper, the relationship between incarceration and crime is complex.

More recent analyses have estimated that while about 25 per cent of the drop in violent crime in the US can be attributed to increased imprisonment, most of the decline was due to other factors - the stronger economy that produced jobs and opportunities for lower-income workers; the sudden end of the crack cocaine epidemic; smarter policing; and community response changes.

There are many reasons to dislike our home-grown version of the three strikes law.

Critics have described it variously as "irrational and extreme", concerning and unjust, as well as plain "stupid" (criminologist Greg Newbold's assessment back in January).

Legal academics Dr Richard Ekins and Professor Warren Brookbanks argued in recent lectures sponsored by the Maxim Institute that the proposed law will fail in its aim to deter offenders, and lead to arbitrary and unfair sentencing.

Worse still is that, rather than targeting "the worst of the worst", it will radically widen the scope of offenders to include those on relatively less serious offences.

In other words, it misses the point.

We all want a safer country, with much less violent crime. We all want our most vicious criminals locked up for as long as it takes to keep us safe. But no law will guarantee that we'll sleep more safely in our beds, or keep our children out of harm's way.

Especially not if we ignore the real drivers of crime, as National seems intent on doing (as the Ministry of Justice has pointed out).

Western Australia's retiring District Court Chief Judge Antoinette Kennedy recently said: "Once you can have people more frightened of disorder than tyranny, it enables you to do almost anything you like so far as legislation is concerned.

"... it doesn't require any leadership to say we're going to increase all penalties and we're going to lock everybody up longer. But to actually convey to the community, no, we're not doing that because it doesn't work, we're going to do these other things such as early intervention, such as programmes - that requires leadership."