Australian Institute of Criminology

Skip to content


Armed robbery is often portrayed as a professionally executed, high-return exercise for the offenders involved. Recent research undertaken in the United Kingdom into the financial benefits of bank robbery—perhaps the most pervasive of those commonly portrayed images of robbery—demonstrates that the gains from this crime are far from exorbitant given the relative risk to offenders. Around one-third of bank robberies that were analysed netted offenders nothing and on average, bank robbers could expect to ‘earn’ less than £13,000 per person per raid. Furthermore, bank robberies were a relatively rare occurrence when considering the large number of robberies that took place in the reference period (see Reilly, Rickman & Witt 2012). It appears that culturally common understandings of robbery are not necessarily borne out by the data.

Accurate information concerning the nature of armed robbery can assist in providing a clearer picture of the offence. This, in turn, will assist in developing appropriate crime prevention responses that are based on a realistic understanding of what armed robberies can entail.

Monitoring armed robbery in Australia

The AIC’s NARMP has been recording and reporting on armed robbery since 2003. The program was established to monitor trends in armed robbery (specifically trends in weapon use), identify changes in trends and provide insight into the factors underpinning these trends. It was initially modelled on the Recorded Crime: Victims, Australia (RCV) collection (eg ABS 2012a) and therefore records information about every victim of armed robbery reported to police in Australia, with police administrative data received from each state and territory.

Stakeholder consultation has seen refinements to what is collated over time. For example, victim data from calendar year 2004 onwards have usually been accompanied by an incident identifier. This allows victim records to be collapsed into the incidents in which those victims were involved. The capacity to analyse data by incident is important for the accurate description of the key elements of armed robberies. For instance, a single armed robbery involving one handgun might have six victims. If data are analysed in a victim-based format, a count of six handguns would result, but if the unit of analysis is the incident, only one handgun is counted, better reflecting the reality of the crime.

Box 1: Victims of armed robbery

For the purposes of NARMP, a victim refers only to those individuals or organisations whose property was the target of the armed robbery. The AIC understands that individuals who may have witnessed the incident or been involved in some way other than via property ownership are clearly victims in the more common use of the term. However, due to NARMP recording practices, these individuals are not included in the following analyses. This report only describes the characteristics of those victims who were involved in armed robbery as owners of the targeted property, not all possible victims of armed robbery.

The level of detail in the collated information has also increased over time. The initial annual dataset mainly contained information pre-coded into higher level RCV categories. Files received from some jurisdictions now contain information in its raw form, which allows more detailed categories to be constructed. However, the small numbers in some categories can vary widely over time due to chance factors, so an apparently large percentage change over time may in reality only represent a small number of cases. This limits the capacity to make reliable yearly comparisons. Some variables are not recorded in NARMP, such as details on sentencing and an offender’s prior convictions. This information can now be found in some jurisdictional reports such as Victoria’s Sentencing Advisory Council (2010) report Sentencing for Armed Robbery: A Statistical Profile.

Additional detail concerning methodology and the type of information included in NARMP can be found in the Technical Appendix to this report, as can a more detailed discussion of the limitations of the NARMP data collection. This Technical Appendix also contains a glossary of terms and definitions relevant to this report.

The structure of this report

To date, armed robbery information has been explored and reported on an annual basis. However, this is the first report produced since the decision was taken to shift to biennial reporting and as such, it provides information on two years of armed robbery data—for the calendar years 2009 and 2010. For the purposes of this report, ‘recent armed robberies’ refers to 10,409 incidents, involving 12,005 victims reported to police between 1 January 2009 and 31 December 2010.

The report is organised into five sections:

  • Key trends in armed robbery between 2003 and 2010.
  • Characteristics of victims and offenders in 2009 and 2010.
  • Physical aspects of recent armed robberies, such as location and time.
  • The objects associated with armed robbery—the weapons used and the types of property taken in 2009 and 2010.
  • A case study exploring patterns and characteristics of the small number of female armed robbers.

The unit of analyses reported shifts between victim and incident, depending on which aspect of armed robbery is being considered. Presented data and accompanying commentary indicate which unit of analysis has been employed. Finally, Tables separately summarising data relating to each of the years 2009 and 2010 can be found in the Technical Appendix.

Related links

Armed robbery in Australia 2009–10: National Armed Robbery Monitoring Program report: