The Vanishing Criminal: Causes of Decline in Australia’s Crime Rate

By Don Weatherburn and Sara Rahman

In 2000, Australia had the highest rate of burglary, the highest rate of contact crime (assault, sexual assault and robbery) and the second highest rate of motor vehicle theft among the 25 countries included in the international crime victim survey, which takes in the United States, the United Kingdom and most western European countries.

Then in 2001, Australian crime statistics began to decline. By 2018, rates of the most common forms of crime had fallen between 40 and 80 percent and were lower than they’d been in twenty or in some cases thirty years. Australia is not the only country to have experienced this social trend.

In The Vanishing Criminal Don Weatherburn and Sara Rahman set out to explain the dramatic fall in crime, rigorously but accessibly comparing competing theories against the available evidence. Their conclusions will surprise many and reshape the terms for discussion of these questions well into the future.

• Language: English
• Publisher: Melbourne University Publishing Ltd
• Release date: Feb 2, 2021
• ISBN: 9780522877359

Book preview

THE VANISHING CRIMINAL

THE

VANISHING

CRIMINAL

CAUSES OF DECLINE IN AUSTRALIA’S CRIME RATE

DON WEATHERBURN AND SARA RAHMAN

MELBOURNE UNIVERSITY PRESS

An imprint of Melbourne University Publishing Limited

Level 1, 715 Swanston Street, Carlton, Victoria 3053, Australia

[e-mail]

www.mup.com.au

First published 2021

Text © Don Weatherburn and Sara Rahman, 2021

Design and typography © Melbourne University Publishing Limited, 2021

This book is copyright. Apart from any use permitted under the Copyright Act 1968 and subsequent amendments, no part may be reproduced, stored in a retrieval system or transmitted by any means or process whatsoever without the prior written permission of the publishers.

Every attempt has been made to locate the copyright holders for material quoted in this book. Any person or organisation that may have been overlooked or misattributed may contact the publisher.

Cover design by Peter Long

Typeset in 12/15pt Adobe Garamond Pro by Cannon Typesetting

Printed in Australia by McPherson’s Printing Group

9780522877342 (paperback)

9780522877359 (ebook)

Contents

Preface

1The Rise and Fall in Australian Crime

2Explanations for the Crime Drop: A First Pass

3Is It All about Demographics?

4The Rise and Decline of the Heroin Gen

5Men, Booze and Violence

6Bars, Locks and Alarms: Are We Just More Secure?

7Shopping for Stolen Goods

8Is It All Down to the Police, Courts and Prisons?

9Tying It All Together

Acknowledgements

Glossary of Technical Terms

List of Tables

List of Figures

Appendix A Changes in Repeat Offending and Offender Age

Appendix B Alcohol-Related Emergency Department Admissions and Assault

Appendix C Petrol Prices and Petrol Theft

Appendix D Clear-up Rates and the Risk of Apprehension

Appendix E Changes in Crime-Prone Age Groups: Major States

Notes

References

Index

Preface

In the early 1990s, much to the surprise of nearly everyone, crime rates in the United States began falling … and kept on falling. Between 1993 and 2017, violent crime (as measured by crime victim surveys) fell by 74 per cent, while property crime fell by 69 per cent. The crime drop was unprecedented and attracted enormous scholarly attention; however, the United States was not alone in experiencing a dramatic fall in crime. Significant reductions in crime have also been recorded in Canada, Australia, New Zealand, the United Kingdom and other parts of Western Europe. There is no scholarly consensus as yet as to what caused the crime drop in the United States, or anywhere else for that matter. Some attribute the decline to the waning of the US crack cocaine market, some attribute it to improvements in household security, while others favour explanations based on the legalisation of abortion (fewer unwanted children), the decline in atmospheric lead levels, or increased rates of imprisonment. At last count, no less than 16 theories have been put forward to explain the crime decline or some aspect of it.

The fact that crime has fallen substantially in a number of countries naturally prompts the thought that there must be a single factor or set of factors operating in each. There are, however, some noteworthy differences between countries in the timing and character of the crime decline. The fall in crime in the United States and Canada affected all major categories of crime and began in the early 1990s. The decline in crime in the United Kingdom was similar in scope but did not begin until the mid-1990s. The decline in homicide in Australia began in the late 1980s, but the property crime drop did not begin until 2001 and the decline in assault did not begin until 2008. The timing of the fall in crime in New Zealand also varied from offence to offence. These differences in the timing and character of the fall in crime suggest to us that there may not be a single explanation that applies to all the affected countries—a suspicion reinforced by the manifest differences between countries in the behaviour of factors identified as possible causes of the fall in crime.

In this book, we take each of the theories that have been put forward to explain the crime decline and assess them against the available evidence. Our assessment is in one sense quite parochial. Although our starting point is with the question of whether a particular explanatory factor has the crime-reducing effects claimed for it, we are ultimately concerned with the question of whether that factor is operating in Australia in a way that would produce a significant reduction in crime. This leads us to dismiss or discount some factors (e.g., right-to-carry gun laws, legalisation of abortion) that might have affected crime in some countries, but which are arguably of no or only limited relevance to Australia. That said, although our focus is on Australia, our inquiry should be of interest to scholars and readers in other countries because it leads us to examine factors (e.g., changes in the market for stolen goods) that have received comparatively little attention despite their potential relevance outside Australia.

It became clear to us during the course of this inquiry that the fall in property and violent crime in Australia could not reasonably be said to be the result of a single factor moving in a propitious direction. The best explanation for the decline in homicide seems to be the steady improvement in emergency medical treatment. The best explanation for the fall in assault, on the other hand, seems to be the decline in alcohol abuse among young people in response to rising alcohol prices. Changes in some factors (e.g., the fall in heroin dependence) had knock-on effects (a reduction in the number of offenders relative to the number of police; a decline in the size of the market for stolen goods) that allowed other factors (increased risk of apprehension) to exert a greater effect. Luck also played a role. Public demands for increased police accountability reached their peak just as knowledge about what works in policing began to filter into policies on police strategy and deployment. All this occurred against a backdrop of low inflation, rising real average weekly earnings and falling unemployment.

Given the unusual combination of conditions that led to the fall in crime in Australia, we remain uncertain about how long it can be expected to continue. Australia at the moment is experiencing two epidemics: one involving methamphetamine use and the other involving the highly infectious disease known as COVID-19. So far, the methamphetamine epidemic does not seem to have had a significant effect on property and violent crime, but that may change as an increasing proportion of users make the transition from recreational to dependent use. The onset of the COVID-19 epidemic has had two effects, both of which have affected or will affect crime. Its more immediate effect has been to reduce social interaction, which has brought about a further fall in property crime. Whether it results in an increase in violent crime and/or unlawful acts such as fraud, cybercrime and family violence remains to be seen. The second effect of COVID-19 is that it has brought Australia’s economic dream run (25 years since the last recession) to an abrupt end. The effects of this will take time to play out.

CHAPTER 1

THE RISE AND FALL IN AUSTRALIAN CRIME

Baby Boomers Cut Loose

On the morning of 31 January 1984, 35-year-old Hakki ‘Tim’ Bahadir Atahan entered the Commonwealth Bank in George Street, Sydney intending to rob the bank, only to find the building surrounded by heavily armed police. A few hours of failed negotiation followed before Atahan forced five hostages into a Datsun sedan and instructed one of them to drive towards Manly, a beachside suburb north of the CBD. The Datsun, followed by a swarm of police cars and a police helicopter, made it to the Spit Bridge where it was forced to stop—police had raised the bridge to prevent any further escape. Detective Senior Constable Steve Canelis, who had been in one of the pursuing police cars, walked over to Atahan hoping to negotiate his surrender. Without warning, Atahan shot him and was then himself shot dead by other police. The hostages and Detective Senior Constable Canelis all survived, although Canelis later left the police with a bullet still lodged in his shoulder (Anderson 2014).

Sydney’s ‘dog day afternoon’, as it later became known, graphically illustrated the crime crisis gripping Australia at the time. It was a period of unrelieved criminal violence. Three months after the shooting, the Parramatta Family Court was bombed, followed by bombings at the homes of two judges, one of which killed the wife of a judge. In September that same year, seven people were killed and 28 injured when two rival motorcycle gangs engaged in a shootout in the car park of the Viking Tavern in the Sydney suburb of Milperra. Eighteen months later, in March 1986, four men planted a bomb outside the Russell Street Police Station in Melbourne, which killed one officer and wounded 22 others when it exploded. The following year, a German tourist named Joseph Schwab shot five people dead at Timber Creek in the Northern Territory before being shot dead himself. In August 1987, a 19-year-old Victorian man named Julian Knight went on a random shooting spree in Melbourne, killing seven people and injuring 19. Two months later, Frank Vitkovik shot dead eight people and injured five others in a Melbourne post office before leaping to his death from an 11th floor window (Anderson 2014). In 1990, Paul Evers shot five people dead in Surry Hills, New South Wales. The following year, Wade Frankum shot seven people dead and injured six others in a mass shooting in Strathfield, New South Wales. Five years later, Martin Bryant slaughtered 35 people and wounded 23 others in a mass shooting in Port Arthur, Tasmania.

Lethal violence was not the only crime problem facing Australia at the time. Home break-ins (aka burglaries), motor vehicle thefts and robberies were rampant. There is no consistent set of recorded crime figures for Australia from the mid-1970s to 2000, but Figure 1.1 shows the trends in recorded crime across Australia between 1973/74 and 2000 using the two available sources.

Although the break in the series and the associated change in counting rules and offence definitions make it impossible to calculate a meaningful change in crime rates, it is obvious that Australia during the 1970s, 1980s and 1990s faced rapidly rising rates of break and enter, motor vehicle theft, robbery, stealing, assault and fraud. And this was not a case of big changes off a low base. When the results of the 2000 International Crime Survey were published, Australia was revealed as having the second-highest rate of car theft, the highest rate of burglary, the highest rate of contact crime (robbery, sexual assault and assault with force) and the highest overall level of crime victimisation (van Kesteren, Mayhew & Nieuwbeerta 2000).¹ At this stage, one in 20 Australian households was falling victim to burglary every year, one in 60 was losing a car to theft, and one in 20 people over the age of 15 was being assaulted (Australian Bureau of Statistics 1999a).

Figure 1.1 Recorded rates of burglary, motor vehicle theft, robbery, stealing, assault and fraud—Australia, 1973/74 to 2000

Sources: Data from Australian Bureau of Statistics (2001a); Australian Institute of Criminology (1990)

No state or territory escaped the rise in crime. Figure 1.2 shows the average annual percentage increase in recorded crime rates by offence type between 1973/74 and 1988/89 for New South Wales, Victoria, Queensland, Western Australia, South Australia and the Australian Capital Territory.² None of the listed states and territories experienced a decline in any of the offences listed. In every jurisdiction, every offence went up—in some cases by a large margin. In South Australia and Western Australia, the recorded serious assault rate between 1973/74 and 1988/89 rose at nearly 14 per cent per annum. In Queensland it rose over the same time period by more than 14 per cent per annum.

It has sometimes been said that the increase in recorded assault reflects nothing but an increase in the willingness to report the offence to police. Crime victimisation surveys provide a more reliable source of information on trends in assault but, due to changes in question wording, it is not possible to determine the trend in the national prevalence of assault using data from crime victim surveys conducted prior to 2008.³ The only consistent source of data we have on assault between 1990 and 2008 is the annual crime and safety survey conducted in New South Wales during these years by the Australian Bureau of Statistics. That survey showed a steady rise in the prevalence of assault in New South Wales between 1990 and 1998 (see Figure 1.3). Apart from an unusually high level of reporting in 1997, the percentage of assaults reported to police remained steady at around 34 per cent between 1990 and 2008 (Australian Bureau of Statistics 1994; 1997; 2001b; 2004b; 2008b). There is not much reason, then, to believe that the growth in assault in Australia was just a case of an increased willingness to report the offence to police.

Figure 1.2 Average annual percentage change in recorded rates of crime by offence type—Australia, 1973/74 to 1988/89

Source: Data from Australian Institute of Criminology (1990)

Some states and suburbs suffered far more than others. The national prevalence rate of break and enter in 1998 was 5 per cent (one in 20 households) but the WA rate in that year was 7.5 per cent (one in 13). The national prevalence of robbery in 1998 was 0.5 per cent (one in 200) but the NSW rate was 0.9 per cent (one in 111). These percentages might seem small, but a 0.9 per cent figure for robbery equates to an estimated 42 500 victims per annum (Australian Bureau of Statistics 1999a).

Figure 1.3 Trends in the prevalence of assault—New South Wales, 1990 to 2008

Source: Data from Australian Bureau of Statistics (2004b; 2006a; 2008b)

The variation in crime risk across suburbs and towns was even starker. In 2000, the recorded NSW break and enter rate was 1357 per 100 000 population. In North Sydney, the burglary rate was 55 per cent higher. In Bourke, the recorded rate of burglary was 2.7 times higher (3628 per 100 000 population) (NSW Bureau of Crime Statistics and Research 2019). Small wonder the NSW town was beset with periodic riots (Weatherburn 2006).

The Public, the Media and the Politicians

The spread of lawbreaking into the suburbs led to rapidly rising public concern, fuelled by an insatiable media and political appetite for stories about rising crime. Between 1991 and 2000, the proportion of NSW residents expressing concern about crime in their neighbourhood grew from 30.6 per cent to 33.0 per cent for home burglary, 18.9 per cent to 21.8 per cent for car theft, 17.1 per cent to 19.5 per cent for ‘louts/youth gangs’, 10.3 per cent to 17.3 per cent for illegal drugs, and 17.4 per cent to 24.2 per cent for vandalism/graffiti (Australian Bureau of Statistics 1991; 2000).⁴ No similar surveys were conducted in other states and territories over the same time period but, since those regions were also undergoing rapid increases in crime, it would be surprising if they were not also experiencing the same growth in public concern about crime. In 2002, the Australian Bureau of Statistics (2003) estimated that 6.3 million Australians (44.9 per cent of the country’s adult population) were concerned about break and enter as a problem in their neighbourhood, while more than one in four (21 per cent, or three million adult Australians) were concerned about car theft in their neighbourhood.⁵

It was not just theft and violence that worried the Australian public. The country at the time was in the grip of a heroin epidemic, which many believed to be responsible for the epidemic of theft and robbery. The proportion of all deaths resulting from opioid overdose among adults aged 15–44 increased from 0.1 per cent in 1964 to 7.3 per cent in 1997. Over the same period, the opioid-caused mortality rate (per million population) in this age group rose from 1.3 to 71.5 (Hall, Lynskey & Degenhardt 1999).

You did not have to be a student of health or crime statistics to know the country had a drug problem. In 1985, a tearful Bob Hawke announced in a nationally televised prime-ministerial address that his daughter was a heroin user and that his government would be undertaking a National Campaign Against Drug Abuse. The ‘Grim Reaper’ television campaign, launched two years later at the height of public concern about the HIV/AIDS epidemic, added to the high level of anxiety about public safety. In the 20-year period between 1977 and 1997, there were three royal commissions into drug trafficking,⁶ a royal commission into drug-related corruption in the NSW Police Service, and a NSW Legislative Council inquiry into why the NSW Police had failed to reduce drug trafficking in the south-western Sydney suburb of Cabramatta, then the heroin-dealing capital of Australia.

Public concern about the risks posed to the community by drug-dependent offenders accelerated as the number of persons dying from AIDS-related diseases increased,⁷ and offenders began using syringes as a weapon in the commission of robberies.⁸ On 31 January 1999, a photograph of a teenage boy injecting drugs in a laneway in Redfern appeared on the front page of a Sydney newspaper, prompting the NSW Government to announce a Drug Summit (Swain 1999). As the drug epidemic progressed, certain suburbs (e.g., Redfern and Cabramatta in Sydney, Richmond and Abbotsford in Melbourne) seemed constantly blighted by public drug use, drug trafficking and discarded drug-use paraphernalia. Cabramatta was described by one senior police officer as a ‘war zone’ at the time (Kwai 2017). A kind of fevered irrationality descended upon politicians, ushering in what Roberts et al. (2003) would later describe as the era of ‘penal populism’—an era in which political parties competed aggressively with one another over who could be tougher on crime.

In 1989, Western Australia had a serious car theft problem. Car thefts in that year were running at nearly 10 per 1000 population, a figure higher than in any other state or territory of Australia, and substantially higher than the corresponding rates being recorded by police in the United States and Britain. Whenever they could, the WA police responded to the problem by engaging in high-speed pursuits of stolen cars. This practice had disastrous consequences. Over an eighteen-month period, from around April 1990, 16 people died as a result of accidents that occurred during such high-speed chases.

As the number of deaths rose, so too did the level of public concern. In August 1991, a public rally held outside the WA Parliament to demand tougher government action against vehicle thieves attracted 20 000 people. Then, on Christmas night that year, a pregnant mother and her baby son were killed when their car collided with a stolen vehicle being chased by police. A month later, the state government promised ‘the toughest laws in Australia’ to deal with crime. Repeat juvenile offenders were to be detained ‘at the Governor’s Pleasure’ (i.e., indefinitely) in addition to any other penalty that was imposed. All crimes involving violence were to be accompanied by a minimum 18-month jail term. Penalties for driving offences resulting in the death of someone were to be increased substantially where the offence involved a stolen motor vehicle. However, the new laws had no effect. The car theft problem was only solved, eventually, with the introduction of engine immobilisers.

The Northern Territory’s mandatory sentencing laws are another example of the political madness that prevailed at the time. They were enacted in March 1997 in response to an alleged epidemic of home break-ins. Under the laws, certain property offences attracted mandatory minimum prison terms, even for first offenders. The relevant legislation was the subject of national and international criticism because of its perceived unfairness. In one case dealt with under the new legislation, a young Aboriginal offender who was an orphan, with a long history of social problems, received a mandatory sentence of 28 days’ imprisonment for stealing stationery with a value of less than $50—the boy hanged himself on the 24th day of his detention (Howe 2001). There is no evidence that mandatory sentencing reduced burglary in the Northern Territory. In fact, the year after the introduction of the mandatory sentencing laws in the Northern Territory, the recorded rate of home break-ins actually went up (Australian Bureau of Statistics 1999a).⁹

In the lead-up to the 1995 Queensland election, the state Labor government and the Liberal–National Opposition began a bidding war on law and order. The government promised to spend $1 billion on a law and order package to cut crime. The package included an additional 2000 operational police staff over the next 10 years (with an extra 500 over the following three years), an extra $54 million to build new prisons, $37 million to build new police stations and watch houses, and $20 million to upgrade courts and Justice Department premises. Not to be outdone, the Opposition promised an additional 2780 additional police over the next 10 years, and new legislation to ensure that serious violent offenders spent 80 per cent of their sentence in prison and which allowed courts to impose more than the statutory minimum sentence of 20 years on multiple murderers. As an added bonus, the Opposition also promised to pass new laws allowing a home occupant, without fear of prosecution, to shoot home invaders.

The media, of course, played a key role in stimulating growing public concern about rising crime. It is difficult to capture the mood of the 1980s and 1990s, but the headlines below were typical.

Sources: The Sydney Morning Herald (1985). ‘Survey: police overwhelmed by crime wave’. 3 November; The Sydney Morning Herald (1994). ‘Judges slam weak laws’. 26 June.

On Sunday 31 May 1992, the then leader of the federal Opposition, Dr John Hewson, was reported in the Sydney Sun-Herald as having said that Australia was the crime capital of the world. Hogg and Brown (1998) captured the mood perfectly in their book Rethinking Law and Order, relating how the country’s only national newspaper, The Australian, claimed in 1985 that Australia had embarked on an ‘orgy’ of violent crime (p. 27). Four years later, The Sydney Morning Herald reached the same conclusion, claiming that there was ‘without doubt, a new climate of violence in Australia, of a kind not experienced by earlier generations’ (Hogg & Brown 1998, p. 27).

Courts and Prisons Feel the Squeeze

Not surprisingly, state and territory governments started spending more on police. Total national expenditure on policing amounted to $2512 million in 1992/93, accounting for about 4 per cent of all state and territory government final expenditure in that financial year. Over the decade 1982–93, real expenditure on policing grew at an annual average of 3.9 per cent (Commonwealth of Australia 1995, p. 366). Between 1994/95 and 1998/99, state and territory government expenditure on law and order increased in real terms from $4.8 billion to $5.9 billion, an increase of 23 per cent in just five years (SCRCSSP 2000, p. 506).

Equally unsurprisingly, arrest rates rapidly increased. Between 1973/74 and 1987/88, the national arrest rate rose 288 per cent for serious assault, 74 per cent for robbery, 52 per cent for break and enter, 17 per cent for motor vehicle theft and 95 per cent for fraud. Between 1974/75 and 1987/88, the recorded rate of drug offences (essentially a record of arrests) rose from 70.7 per 100 000 population to 290.3 per 100 000 of population—a more than fourfold increase (Australian Institute of Criminology 1990).

The combination of rising arrest rates and tougher law and order policies led to congested courts and overcrowded prisons. In 1988, after a sustained period of serious problems with court delays (Callinan 2002), the newly elected Greiner Liberal government hired consultants WD Lybrand to investigate the congested state of the NSW criminal court system. Australia, meanwhile, was experiencing an unprecedented growth in prison numbers. In the 66 years between 1918 and 1984, the Australian imprisonment rate rose by just 13 per cent (Australian Institute of Criminology 1988), but in the 16 years that followed, it rose by more than 83 per cent (Australian Bureau of Statistics 2008a; 2018a; Carlos & Grant 1999). It must have seemed to most people that Australia was permanently enmeshed in a world of rising crime, tougher law and order policies, and increased expenditure on dealing with wrongdoing.

Theft and Robbery Start to Tumble

But it was not. New South Wales was among the first states to see the change. Early in 2001, recorded rates of robbery and break and enter in New South Wales unexpectedly began falling. Figures 1.4 to 1.7 show, respectively, the monthly number of robbery, burglary, motor vehicle theft and stealing from a motor vehicle offences in New South Wales between January 1999 and November 2002. The data have been smoothed to remove the random monthly fluctuations and make the trends easier to see.¹⁰

No-one saw the change coming—not the public, not the media commentators, not the police, not even the crime statisticians or the criminologists. The trends were so sudden and so unexpected that some in the media simply refused to believe them. Conservative political commentator Miranda Devine blamed the drop in reported crime on declining public willingness to report crime to police, claiming that it was caused by the introduction of the new police assistance line