Effective law and order policy need not be a shot in the dark

Don Weatherburn | SMH | January 10, 2012

It's hard to find anyone these days who doesn't believe in evidence-based policy. Politicians swear their allegiance to it, bureaucrats praise it and I've even been lectured by a taxi driver on the need for a more evidence-based approach to crime control.

Evidence-based policy stands in contrast to policy based on hunch, prejudice, guesswork or the desire for a ''warm inner glow''. The public service's embrace of it has been more rapid and fervent than Saul's conversion on the road to Damascus. If this epiphany were real, ordinary taxpayers would have much to celebrate, especially in the domain of law and order. But, sadly, it can't be believed.

You would never be able to market a pharmaceutical drug in Australia without rigorous evaluation by the Therapeutic Goods Administration. But state and territory governments routinely spend large sums of taxpayers' money trying to reduce crime and re-offending without subjecting the measures to any evaluation. Where evaluations are undertaken, the results are often ignored.

The promise to appoint additional police and impose tougher penalties on crime are staples at nearly every election; yet no Australian state or territory government has ever promised to evaluate and publicly report on the effects of additional police and tougher penalties.

And it isn't just those old staples that escape critical scrutiny. The list of policies shown by my office to have no effect on re-offending in NSW includes high fines for drink drivers, supervision of offenders on good behaviour bonds, detention for juvenile offenders, the forum sentencing program (a restorative justice program for young adult offenders) and the circle sentencing program (under which Aboriginal offenders are brought before community elders for sanctioning).

Despite the negative results, all these policies remain in place. Meanwhile, programs that have been known for years to be effective, such as the NSW Drug Court Program, are only now being expanded.

Why do governments trumpet the virtues of evidence-based policy, while often ignoring it in practice? One reason is that law and order policy is as much, if not more, influenced by what's popular than by what's effective.

Tough penalties and measures that give victims an opportunity to confront offenders and demand an apology and restitution go down well with the general public. Giving cognitive behavioural therapy to violent offenders to help them manage their anger is nowhere near as attractive, even though it works.

Another reason is that, far from being politically neutral, senior public servants often have political or policy agendas of their own that they want to pursue.

A third reason is that unlike areas such as those of health and education, criminal justice agencies have no entrenched culture of research and development.

Fourth, the public servants who end up managing rehabilitation programs often end up becoming forceful advocates for those programs and trenchant critics of evaluations that produce negative results.

The long-term solution to all this is a better-informed public and more rigorous scrutiny of law and order policy by the media. Here are 10 questions one should ask of any government that declares its commitment to evidence-based policy:

1. Does the government state the objectives of its law and order policies and programs in terms that can be measured? If not, there is no way they can be properly evaluated.

2. Does the government base its policies and programs on the results of systematic reviews, such as those published on the website of The Campbell Collaboration, an international research network? These reviews objectively summarise the results of all past rigorous research into the effectiveness of various interventions in preventing crime and reducing re-offending.

3. Are the government's law and order policy advisers trained in both research methods along with statistical analysis?

4. Does the government provide researchers with comprehensive access to information on the rate at which convicted offenders are reconvicted?

5. Does the government provide to researchers comprehensive access to all information on reported crime?

6. Are all major new programs subjected to rigorous cost-benefit or cost-effectiveness evaluation by an independent agency?

7. Are all evaluations subjected to independent peer review by appropriate experts in the field to detect flaws?

8. Does the government abandon or substantially modify programs that have been shown to be ineffective in achieving their stated goals?

9. If the government substantially amends a policy or program, is the revised policy/program evaluated?

10. Does the government ever delay or withhold the results of evaluations it commissions? If so, then the government is clearly keen to persist with policies that are not supported by evidence.

Without this planning, training, transparency and responsiveness, policies may amount to a complete waste of public money.

Where governments are truly committed to evidence-based policy, it should be possible to answer all these questions with a ''yes''.

Dr Don Weatherburn is director of the NSW Bureau of Crime Statistics and Research.

Head Case

By Christopher Beam, Slate, February 11 2011:

Crime rates have plummeted over the last 20 years. Why aren't we less scared?

When my grandfather learned I was going to college in New York City, he knew what to give me for a high-school graduation present: a hunting knife. If I was going to make it to class, he explained, I would probably have to defend myself. Never mind that this was 2002, and crime in New York City had plummeted since its high in the early 1990s. You could never be too careful.

His concern highlights the bizarre way we think about crime. Even as crime rates have gone down around the country over the last 20 years, our fear of crime hasn't changed much at all. Between 1990 and 2009, the national violent-crime rate was halved, while property crime dropped to 60 percent of its previous rate, according to the National Archive of Criminal Justice Data. But almost every year since 1989, most Americans have told pollsters they believe crime is getting worse.

The disparity has been especially clear in New York City. That city saw the most dramatic crime decline of all: Since 1990, the homicide rate has dropped 82 percent, robbery by 84 percent, rape by 77 percent, and auto theft a stunning 94 percent, according to the New York Police Department. These numbers have now been confirmed in an independent study by Frank Zimring of the University of California, Berkeley, who confirms that the drop in crime is "real."

But New Yorkers don't feel correspondingly safer. "If you walk these streets, especially at night, you know crime is definitely not down," a cab driver who lives in Harlem told the New York Daily News in 2009. "It's not safe. I don't know where they get these statistics." Same with Baltimore, where a statistical decline in crime is no match for the occasional sensational murder that reinforces the city's reputation as a dangerous town.