Australian Institute of Criminology

Skip to content

Illicit market firearms and organised crime

There has been considerable speculation in the public sphere, particularly in response to apparent increases in drive-by shootings and other gang-related shooting offences, on the nature of the illicit firearms market in Australia, specifically the sources of these firearms. However, little formal examination of what this market comprises, how it is replenished and its relationship to SOCG has been available to test this speculation. This lack of analysis is partly due to universal difficulties in quantifying and describing illicit good markets, particularly in the absence of comprehensive information sources.

The best available data to assist in the construction of the illicit firearms market in Australia is that compiled by the ACC on seized firearms. Using analysis of data from the ACC’s NFTD, this section describes the characteristics of firearms found in the illicit market, where these firearms originated and the means by which these firearms ended up in the illicit market. The section focuses on firearms acquired by SOCG (see Table 1 for a definition of serious and organised crime), the prevalence of prohibited firearms in SOCG caches and whether similar patterns of supply to the illicit market are used for restricted and non-restricted models.

General firearm characteristics

A total of 2,750 seized firearms were recorded in the NFTD as of March 2012 (see Methods in first section on the compilation of this data). Where information was recorded on the date of seizure (n=2,341), all but 10 were seized between June 2002 and October 2011. Of the 10 that were recovered earlier, one was seized in 1977 and the others between 1995 and 1999.

Of these seized firearms, 43 percent (n=1,184) were rifles, 34 percent (n=960) were handguns and 16 percent (n=448) were shotguns (see Table 7). Only a small number of prohibited machine gun models have been recorded, comprising less than one percent (n=26) of all seized firearms. Some of these firearms were seized as part of multiple-firearm recovery events, but the quality of the data precluded determining how many firearms were seized individually or as part of a larger assemblage and what these multiple seizures consisted of. The largest number of firearms seized as a collection was 102, recovered in New South Wales from individuals involved in firearms trafficking. Other larger seizures associated with SOCG included the recovery of 85, 45 and 35 firearms, all from entities involved in the illicit drugs market, and 60 firearms from a firearms trafficking venture. There was a small group of large seizures from non-SOCG too—55 unregistered long-arms from a licensed firearm owner in New South Wales and seizures of 21 and 22 grey market-sourced long-arms from individuals in Queensland.

Table 7: Firearm type seized from SOCG and non-SOCGa
Firearm typen%n%n%
Rifle 672 40 512 49 1,184 43
Shotgun 278 16 170 16 448 16
Air rifle 48 3 58 6 106 4
Handgun 665 39 295 28 960 34
Other 26 2 10 1 36 1
Sub-machine gun 16 1 3 <1 19 <1
Light machine gun 4 <1 2 <1 6 <1
Heavy machine gun 0 0 1 <1 1 <1
Combination firearm 6 <1 4 <1 10 <1
Unknown 12 1 4 <1 16 1
Total 1,701 1,049 2,750

a: Percentages may not total 100 due to rounding

Source: ACC National Firearm Trace Database

Similar proportions of rifles (40%) and handguns (39%) were recorded from SOCG seizures, while in non-SOCG seizures, rifles were significantly more commonly recovered (and hence it can be assumed more commonly acquired) than handguns (49% of all firearms seized compared with 28% respectively; χ2 =35.26 df=2 p<0.01). SOCG and non-SOCG seizures contrasted solely in the prevalence of handguns, with a significantly greater proportion of handguns found in association with SOCG.

Both SOCG and non-SOCG firearms were disproportionately skewed towards restricted firearm categories (ie Category C, D and H firearms as classified in the National Firearms Agreement (1996); see Figure 1 and Table 8). Category C and D long-arms comprise self-loading (ie semi-automatic and automatic) rifles and pump action shotguns that were subject to the 1996 gun buybacks and Category H comprise handguns. Altogether, Category C, D and H firearms make up less than 10 percent of all registered firearms in Australia but comprised over 50 percent of all seized firearms. This skew towards restricted models was significantly more marked among firearms seized from SOCG, where 61 percent of all seized firearms were Category C, D or H compared with 44 percent of non-SOCG firearms (χ2=78.2 df=2 p<0.01).

Figure 1: Firearm category (%)a

 Figure 1: Firearm category

a: Excludes unknown category (n=16)

Note: Percentages may not total 100 due to rounding

Source: ACC National Firearm Trace Database

Category H handguns comprised the largest proportion of restricted firearms in both SOCG and non-SOCG seized firearms but, as described earlier, were significantly more prevalent among firearms recovered from the former group. There was little difference in the percentage of Category C firearms between SOCG and non-SOCG but the proportion of Category D firearms seized from SOCG (15%) was almost double that of non-SOCG firearms (9%).

Thirty different firearm types were seized from SOCG and non-SOCG alike and, while there were similarities in the predominance of specific firearm types in both groups, the proportional composition was significantly different between the two (χ2=135.13 df=2 p<0.01). Just over a quarter (26%, n=436) of all SOCG-seized firearms were semi-automatic pistols and 15 percent each were bolt action rifles (either Category A or B, n=255) and restricted semi-automatic rifles (either Category C or D, n=253; see Figure 2). Semi-automatic pistols and semi-automatic rifles were also among the more common firearms seized in non-SOCG contexts, making up 18 percent (n=191) and 11 percent (n=116) of all non-SOCG firearms (see Figure 3). Bolt action rifles, the most widely held rifle type among legal owners, were the most common firearm type recovered from non-SOCG (21%, n=217).

Figure 2: Firearm action typea—SOCG (%)b

 Figure 2: Firearm action type

a: PSA=semi-automatic pistol; RBA=bolt action rifle; RSA=semi-automatic rifle; REV=revolver; SSB=single barrel shotgun; RSS=single shot rifle; SPA=pump action shotgun; RLA=lever action rifle; SDB=double barrel shotgun

b: Percentages may not total 100 due to rounding

Note: Excludes unknown action type (n=12)

Source: ACC National Firearm Trace Database

Figure 3: Firearm action typea—non-SOCG (%)b


a: RBA=boligure 3: Firearm action typet action rifle; PSA=semi-automatic pistol; RSA=semi-automatic rifle; RSS=single shot rifle; REV=revolver; SSB=single barrel shotgun; AIR=air rifle; RLA=lever action rifle; SDB=double barrel shotgun

b: Percentages may not total 100 due to rounding

Note: Excludes unknown action type (n=4)

Source: ACC National Firearm Trace Database

Defacement or obliteration of serial numbers is used to conceal the identity of a firearm (eg if used to commit a violent crime or stolen from a victim of violent crime such as armed robbery) and disguise the method of diversion. A total of 542 firearms or a fifth of all firearms seized were recorded as having the serial number defaced. Three-quarters of these were handguns, possibly reflecting the long-prescribed legal requirement for handgun registration and hence the impetus to conceal the identity of items leaving the licit market. Although the difference was not statistically significant, of note is that the larger percentage of firearms (53%) with defaced serial numbers was seized from non-SOCG.

Other, typical modifications come in the form of shortening or converting long-arms to produce a handgun-like model. Around one in 10 (9%) of seized long-arms had undergone a category change (to Category H), the overwhelming majority of which (77%) were found in the possession of SOCG. When it can physically be achieved, shortening the barrel and butt stock of a firearm makes it easier for criminals to conceal it in the commission of crimes. One seized semi-automatic rifle had been modified to a Category R firearm as classified under the Weapons Categories Regulation 1997 (Qld). While the specifics of this conversion were not available in the data, Category R weapons include fully automatic machine or submachine guns.

Restricted firearms

Through the National Firearms Agreement (1996), states and territories amended their firearms legislation to restrict the importation and use of military-style automatic and semi-automatic firearms to designated occupational and official purposes. Firearms now restricted are:

  • self-loading automatic or semi-automatic rimfire rifles;
  • self-loading automatic or semi-automatic centre fire rifles;
  • self-loading shotguns; and
  • pump action shotguns.

The National Firearms Agreement (1996) was accompanied by a 12 month firearms amnesty and compensation scheme whereby owners and dealers were compensated for the surrender of newly restricted firearms. Approximately 642,000 firearms were surrendered during this period.

New restrictions around the ownership and use of handguns were brought in with the National Handgun Control Agreement (2002). Restricted handguns were any model that had:

  • a calibre greater than .38”; or
  • a minimum barrel length of less than 120mm for semi-automatic handguns or less than 100mm for revolvers and single shot pistols; or
  • a magazine capacity of greater than 10 rounds.

Approval for handguns with a calibre of .45” may be granted for use in specialised accredited sporting events. A six month nationwide handgun buyback was held between 1 July 2003 and 1 January 2004 to primarily compensate owners of registered handguns rendered restricted by the new laws. An amnesty was run concurrently for unlicensed owners or owners of unregistered handguns.

Restricted long-arms and handguns

Restricted long-arms are defined here as any long-arm denoted in the NFTD as being subject to the 1996 buyback. Altogether, 529 or 30 percent of all seized long-arms recorded in the NFTD were restricted long-arms. The majority of these restricted long-arms were Category D firearms (63%, n=335), which are prohibited under Australian law except for official purposes (mostly related to animal control and welfare). Semi-automatic rifles comprised 70 percent (n=368) of these restricted long-arms and pump action shotguns a much less prevalent 18 percent (n=95; see Figure 4).

Figure 4: Firearm action type of restricted long-armsa (%)

 Figure 4: Firearm action type of restricted long-arms

a: RSA=semi-automatic rifle; SPA=pump action shotgun; SSA=semi-automatic shotgun; SMG=submachine gun; Other includes select fire rifle, light machine gun and heavy machine gun

Source: ACC National Firearm Trace Database

Restricted long-arms were predominantly associated with SOCG—70 percent (n=368) of all restricted long-arms were seized from entities associated with serious and organised crime. The composition of restricted long-arm types in SOCG and non-SOCG seizures was comparable and the majority of restricted long-arms (mostly semi-automatic rifles) were highly restricted Category D firearms, but the proportions of these were significantly greater among SOCG seizures (67% compared with 55%; see Table 9). Two-thirds (77%, n=194) of semi-automatic rifles seized from SOCG were classified as Category D firearms compared with 62 percent (n=72) of non-SOCG semi-automatic rifles.

Table 9: Firearms by category and action type of restricted long-arms by SOCG status
C 122 33 72 45
D 246 67 89 55
Total 368 161
Action typean%n%
RSA 253 69 115 71
SPA 64 17 31 20
SSA 26 7 6 3
SMG 16 4 3 2
RSF 5 1 3 2
LMG 4 1 2 1
HMG 0 0 1 1
Total 368 161

a: RSA=semi-automatic rifle; SPA=pump action shotgun; SSA=semi-automatic shotgun; SMG=submachine gun; RSF=select fire rifle; LMG=light machine gun, HMG=heavy machine gun

Note: Percentages may not total 100 due to rounding

Source: ACC National Firearm Trace Database

Restricted handguns are defined as any handgun denoted in the NFTD as being subject to the 2003 handgun buyback (ie they had a calibre greater than 38”, a barrel length shorter than the length prescribed and/or a magazine capacity greater than 10 rounds). Compared with long-arms, a much higher proportion of recovered handguns were restricted models (65%, (n=631) compared with the 30% for long-arms). Most (68%) of these 631 restricted handguns were seized from SOCG. Semi-automatic pistols were favoured by SOCG and non-SOCG alike, making up 72 and 74 percent respectively of seized restricted handguns (see Table 10). Revolvers comprised around a fifth of restricted handguns for SOCG, as they did for non-SOCG.

Table 10: Firearm category and action type of restricted handguns by SOCG statusa
Action typeSOCGnon-SOCG
PSA 311 72 148 74
REV 95 22 39 20
PSS 13 3 4 2
BPR 7 2 3 2
Other 5 1 6 3
Total 431 200

a: Percentages may not total 100 due to rounding

Note: PSA=semi-automatic pistol REV=revolver PSS=single shot pistol BPR=black powder revolve Other=air pistol, black powder pistol, derringer, double barrel pistol, multi barrel pistol

Source: ACC National Firearm Trace Database

Another way of differentiating the acquisition of restricted firearms by SOCG and non-SOCG entities is to compare the proportions these firearms represent in the individual firearm pools. With regard to handguns, around two-thirds of all handguns seized from SOCG were restricted forms (65%) as was the case for handguns seized from non-SOCG (67%). This suggests that, while most restricted handguns were associated with SOCG, as stated above, either there was no overwhelming predilection for restricted models by SOCG entities or access to restricted models was equally open to both SOCG and non-SOCG buyers. However, a different pattern emerges with long-arms. Over a third (36%) of all SOCG long-arms were restricted models, significantly higher than the 21 percent found for non-SOCG long-arms (χ2=43.3 df=2 p<0.01).

Restricted ammunition

As noted under Methods, attempts were made to acquire data on the import of 25 ACP, 32 ACP and 380 ACP ammunition, calibres of ammunition that can only be used in SPPs, a restricted handgun model that is attractive to criminals due to its small size. SPPs are also manufactured in other calibres but as ammunition for these SPPs can be used in other firearms (such as rifles) the actual quantity of ammunition associated with the use of SPPs cannot be identified. A total of 143 SPPs chambered for these calibres were seized, 63 percent (n=90) of which were denoted as SOCG seizures. Issues with the quality of import and registration data collected in the NFTD (see below) prevents determining when these SPPs entered the country (ie before or after the 2002 handgun reforms) or whether the import was legal or not.

An examination of quantities of ammunition seized by police (n=62,133 rounds) found the most common ammunition calibres recovered were .22 (rimfire; 44% n=27,587), followed by .30 calibre (7.62mm; 13% n=8,211), .38 calibre (9mm; 12% n=7,597) and 12 Gauge (10%, n=6,359). Of the .32 calibre ammunition seized (n=2,065; 3% of all ammunition seizures), 63% was restricted 32 ACP. Almost all of the .25 calibre ammunition, which made up just one percent (n=838) of all ammunition seizures, was restricted calibre 25 ACP (90%). Of the 27 cases that involved the use or ownership of a pistol chambered for one of these calibre, 10 were seized from individuals charged with drug offences, another 10 for the commission of a violent crime (homicide and armed robbery), six from individuals involved in the supply of a prohibited firearms and one for a drive-by shooting.

Firearms among outlaw motorcycle gangs

OMCGs are involved in a variety of illicit markets, including the stockpiling and trafficking of illicit firearms (ACC 2011, 2008). Just 218 of the illicit firearms recorded in the NFTD were recovered from OMCGs, 13 percent of all SOCG firearms and eight percent of all seized firearms recorded in the NFTD.

Handguns were more common among OMCG-recovered firearms (55%) than among firearms recovered from SOCG in general (39%). Semi-automatic pistols were not just the handgun of choice but the firearm of choice for OMCGs— 40 percent of the firearms recovered from OMCGs were semi-automatic pistols. Semi-automatic rifles and revolvers each comprised less than half the number of semi-automatic pistol numbers seized from OMCGs. Eighty-two percent (n=73) of these semi-automatic handguns were restricted models, significantly higher than the proportion found for SOCG more generally and non-SOCG. Long-arms were correspondingly a less prevalent item (45%) but 50 percent of these were restricted models.

Source of illicit firearms

The grey market, as described earlier, comprises long-arms that should have been registered or surrendered, depending on the restricted status of the firearm, following the 1996 firearm reforms. Grey market firearms were the main source of both restricted (92%) and non-restricted (86%) long-arms (see Figures 5 and 6). Where recognised forms of diversion had been identified, theft was the most common method of transfer, although accounting for just 10 percent of non-restricted long-arms and four percent of restricted long-arms seized from the illicit market. Other methods of supply included illicit domestic manufacture, false deactivation, failure to notify interstate transfer of a long-arm and illegal import—but only for a few of the seized long-arms recorded in the NFTD.

Figure 5: Source or method of diversion for restricted long-arms (%)

 Figure 5: Source or method of diversion for restricted long-arms

a: Other includes deactivation, failure to notify interstate transfer, illegal import, diversion by reporting false loss and serial number transfer (n=8)


Note: Excludes unknown source or method of diversion (n=62)

Source: ACC National Firearm Trace Database

Figure 6: Source or method of diversion for non-restricted long-arms (%)

 Figure 6: Source or method of diversion for non-restricted long-arms

a: Other includes failure to notify interstate transfer, diversion by reporting false loss and nfa (n=9)


Note: Excludes unknown source or method of diversion (n=158)

Source: ACC National Firearm Trace Database

The data on the source or method of diversion for restricted and non-restricted handguns returned very high unknown responses rates (70% and 68% respectively). This is problematic on two levels:

  • the relative importance of trafficking pathways described below may be skewed, producing an over-or underestimation of probable supply routes; and
  • it emphasises where there has been a failure to record or retain relevant tracing information. Some degree of caution is hence required when interpreting this data.

The sources of restricted handguns, and the means by which they were trafficked, stand in contrast to those found for long-arms and reveal the role exploitable legislative provisions had in facilitating the transfer of handguns into the illicit market. False deactivation (39%) and theft or loss of (31%) were the primary sources of restricted handguns that had entered the illicit market where a method of diversion was known (see Figure 7). Other less common forms of diversion collectively made up around a fifth of all seized restricted handguns and included illicit domestic manufacture (mainly of single shot pen guns), dealers failing to record the receipt of a handgun or diverting handguns through false export claims and illegal import.

Figure 7: Source or method of diversion for restricted handguns (%)

 Figure 7: Source or method of diversion for restricted handguns

a: Other includes diversion by false theft, diversion by false frame, failure to notify interstate transfer, serial number transfer, theft staged, diversion nfa and information pending (n=17)


Note: Excludes unknown source or method of diversion (n=301)

Source: ACC National Firearm Trace Database

Theft or loss, rather than false deactivation, was the primary method of supply for non-restricted handguns—50 percent (n=69) of all non-restricted handguns were items stolen from legal owners (see Figure 8). Just over a fifth (21%, n=29) of non-restricted handguns were displaced to the illicit market by reactivating inadequately deactivated handguns. This difference in diversion methods for restricted and non-restricted handguns was significant (χ2=909.5 df=2 p<0.01).

Figure 8: Source or method of diversion for non-restricted handguns (%)

 Figure 8: Source or method of diversion for non-restricted handguns

a: Other includes diversion false export, diversion by spare frame, failure to notify disposal or false disposal notice, failure to notify interstate transfer, illegal import, false loss claim and diversion nfa, (n=24)


Note: Excludes unknown source or method of diversion (n=136)

Source: ACC National Firearm Trace Database

Illicit link

Illicit link data refers to the criminal entity or activity in which the firearm was seized. Any firearm seizure from an individual or group involved in the illicit drug market and/or firearm trafficking, or associated with an organised criminal entity (such as OMCGs), was assigned to SOCG, based on the definition used by the ACC in compiling the NFTD. Other matters, such as seizures of firearms following an incident of violent crime, are assigned to SOCG or non-SOCG depending on the identity or activities of the individuals or entities involved. Illicit link data is not directly comparable between SOCG and non-SOCG.

There was a significant difference in the seizure circumstances for restricted long-arms compared with restricted handguns (χ2=365.7 df=2 p<.01). Of the 368 restricted long-arms recovered from SOCG, 41 percent were seized from entities involved in firearm trafficking, 34 percent from persons involved in the illicit drug market and 13 percent were seized from members of OMCGs (see Table 11). Less than 10 percent were used in the commission of a violent crime. Restricted handguns were mostly seized from persons or groups involved in the illicit drug market. The proportion of handguns seized from persons involved in the drug market was almost double that seized from OMCGs (21%) and persons engaged in firearm trafficking ventures (22%).

Table 11: Illicit link for restricted long-arms and handgunsa
Restricted long-arms
Firearm trafficking 150 41
Drug 125 34
OMCG 49 13
Violent crime 24 7
Firearm offences 12 3
Illegal import 2 1
Other 6 2
Total 368
Restricted handguns
Drug 165 39
Firearm trafficking 92 22
OMCG 89 21
Violent crime 40 9
Firearm offences 27 6
Illegal import 3 1
Other 11 3
Total 427

a: Percentages may not total 100 due to rounding

Source: ACC National Firearm Trace Database

Last known registration status

Detail on the last known registration status of a firearm, combined with data on the location of firearm recovery, can provide information on the transfer of firearms before they are recovered by law enforcement agencies. Pierce et al. (2004) used United States Alcohol, Tobacco, Firearms and Explosives data to do just this, although the data includes records of sales from dealers rather than registration status. In their analysis, Pierce et al. (2004) attempted to calculate:

  • the proportion of recovered firearms that were still in the possession of the original purchaser;
  • the location of first sale and recovery (evidence for jurisdiction transfer); and
  • the length of time between first sale and recovery (time-to-crime).

Unfortunately, the absence until recently of a systematic method of recording Australian firearm imports and domestic sales/transfers and the mandatory registration of all firearms means the data collated on last known registration status not only contains a large number of unknown responses but may not always represent the actual last legal ownership of a firearm before it was diverted into the illicit market. The findings described below can therefore only suggest the possible point or penultimate point of diversion.

From the results presented in Tables 12 and 13, where information was recorded, the last known registration status for the majority of restricted long-arms and handguns was with an Australian dealer, either at import or registered as stock (65% and 56% respectively). By contrast, the last known registration status for non-restricted long-arms was comparably divided between Australian dealers (36%) and individual licence holders (39%). This was not the case for non-restricted handguns which, like restricted handguns, were more likely to have been last registered with an Australian dealer than a private owner (47% compared with 27%). It is difficult to determine whether these findings suggest there was a genuine risk of diversion of restricted firearms by some dealers or whether they are an artefact of previous issues with sales and registration records.

Table 12: Last known registration status for restricted and non-restricted long-armsa
Restricted long-armsnon-restricted long-arms
Australian dealer at import 140 53 124 26
Australia dealer stock 32 12 46 10
Australian individual licence 25 10 187 39
Local commercial manufacture 27 10 96 20
Other 9 3 6 1
Australia (all) 233 89 459 95
Overseas dealer 8 3 5 1
Overseas manufacture 18 7 15 3
Other 3 1 2 <1
Overseas (all) 29 11 22 5
Total 262 481

a: Percentages may not total 100 due to rounding

Note: Excludes unknown=1,023

Source: ACC National Firearm Trace Database

A difference also existed between long-arms and handguns in the site of the last registration, with a larger proportion of handguns having a last known registration with an overseas dealer or manufacturer. This was the case for both restricted and non-restricted handguns.

Table 13: Last known registration status for restricted and non-restricted handgunsa
Restricted handgunsnon-restricted handguns
Australian dealer at import 23 11 18 11
Australia dealer stock 96 45 58 36
Australian individual licence 30 14 43 27
Other 9 4 14 9
Australia (all) 158 74 133 83
Overseas dealer 18 9 5 3
Overseas manufacture 33 16 20 12
Other 4 2 4 2
Overseas (all) 55 26 29 18
Total 213 162

a: Percentages may not total 100 due to rounding

Note: Excludes unknown=582

Source: ACC National Firearm Trace Database


The results presented here provide an indication of the make-up of the illicit firearm market and the suite of firearms held by SOCG and other consumers of illegal firearms. It is suggested by these findings that a combination of preference, availability and connections determines the composition of firearms accumulated.

A preference for restricted models

Not unexpectedly, a high proportion of firearms recovered from SOCG were restricted models— 47 percent of all firearms retrieved from these groups were subject to either the 1996 or 2003 buybacks, compared with 34 percent of firearms recovered in non-SOCG circumstances. Restricted handguns were particularly prevalent. While handguns comprised 34 percent of all firearms seized, restricted handguns accounted for over half (54%) of all restricted firearms recovered. Overall, 65 percent of all handguns found in association with SOCG were restricted models, as were three-quarters of all semi-automatic pistols.

While the majority of firearms recovered from SOCG were in fact long-arms, the apparent preference for handguns is related to their favoured use, according to overseas research, as both a means of protection and in the commission of crime. Data on the use of firearms in the commission of violent crime indicates such a preference (eg Bricknell 2008b; Borzycki 2008; Smith, Dossetor & Borzycki 2011; Smith & Louis 2010, 2009; although there has been a sharp drop in handgun-perpetrated homicides since 2007–08: Chan & Payne (forthcoming); Dearden & Jones 2008). The types of handguns, especially the restricted models, recovered from SOCG have the dimensions and characteristics which most suit SOCG activities. Some semi-automatic pistols and revolvers with 2–3” barrels are concealable and easily carried, an important feature cited in interviews with criminal owners of handguns (eg see Blumstein 1995; Hales, Lewis & Silverstone 2006). Semi-automatic pistols are additionally attractive because their magazines can be easily and quickly changed, and even concealable versions have large magazines capacities of some 10–13 rounds. Concealable revolvers generally have a smaller magazine cartridge capacity (6–5 rounds) and are difficult to reload quickly under stress, which makes them less attractive for use in confrontation episodes. It is concluded in the available literature that handguns are favoured by criminal gangs, or at least by those involved in particular criminal activities such as the illicit drugs market, primarily for self-defence and protection purposes (Blumstein 1995; Bricknell 2008b; Cook et al. 2006; Hales, Lewis & Silverstone 2006; Lizotte et al. 2000). Status is another influential factor, although this is more likely for younger or more impressionable gang members.

Long-arms recovered from SOCG were mostly less restricted Category A and B models (39%) but the proportion of highly restricted Category C and D models (22%) was much higher than is found among registered long-arms. Around a quarter of all SOCG-recovered firearms were Category A firearms and their significance here is probably attributable to the fact that this category of firearms is widely available and hence easily sourced. Category D firearms were of a similar proportion to Category B firearms (15% and 14% respectively) despite being a much less common item in the legal market.

Category C and D firearms tend not to be used in the commission of violent crime and related offences to the same extent as handguns, and their purchase by SOCG is unlikely to be related to the sorts of reasons for which handguns are acquired. Instead it is possible that Category C and D firearms—semi-automatic rifles, semi-automatic shotguns, machine guns etc—are amassed partly due to the ‘attraction’ of owning highly lethal firearms but also to form a cache of firearms that can be drawn upon if and when there is a serious or rapid escalation in animosity between rival groups.

An interesting finding from the preliminary analysis was the difference in the relative proportion of restricted long-arms and handguns between SOCG and non-SOCG. It is assumed that criminal entities are more inclined to possess restricted firearms because they are perceived as the best tool to both protect assets and deter assault. Criminal entities are also much likelier to have established connections with, or operate within, groups that are involved in illicit firearm markets and hence have access to a wider selection of items. Restricted long-arms made up just over a third (36%) of all long-arms seized from SOCG but the proportion was significantly higher than that found for non-SOCG seized long-arms (21%). This was not so for handguns. While the majority of restricted handguns were recovered from SOCG (ie 68%), the proportion of handguns seized from just SOCG that were a restricted model was the same as that proportion found for handguns retrieved from non-SOCG (ie just over two-thirds).

The high concentration of restricted handguns in the non-SOCG pool is probably the result of persons acquiring handguns to suit a curiosity rather than a criminal need. Regulations on handgun use have always been stricter than those for (most) long-arms, and handguns could only be obtained if a person was granted formal membership of a pistol club. Restricted models hence became a coveted item among enthusiasts who may have always wanted a handgun but could never legally obtain one (Project stakeholder personal communication 30 November 2011). The difference then was that handguns were more freely ‘available’ than they are now, an availability that was facilitated by previous state laws regarding the definition and accountability of handguns.

Methods of diversion

In 2000 and 2001 respectively, amendments were made to the Weapons Act 1990 (Qld) and Firearms Act 1996 (NSW) to close loopholes which inadvertently facilitated the diversion of firearms into the illicit market. The Queensland deactivation loophole, described in the second section of this report, was ‘open’ for at least a decade and almost certainly led to the transfer of possibly thousands of handguns from the licit to the illicit market. The Police Powers and Responsibilities and Another Act Amendment Bill 2000 subsequently amended the definition of a firearm to include ‘any Category H weapon that is permanently inoperable’, introduced registration requirements for any Category H firearm (ie handgun) rendered inoperable and prescribed the requirement that a collector’s licence be acquired if a person possesses a permanently inoperable Category H firearm.

The Firearms Act 1996 (NSW) as originally enacted required firearm barrels, but not frames or receivers, to be registered under Part 3 (Registration of Firearms) of the Act. The exemption of frames and receivers meant handguns without barrels could be sold without having to observe regulations on firearm disposal and frames/receivers could be purchased without need to register them. This opened up opportunities to convert or build up new handguns using non-registrable parts purchased in New South Wales with parts purchased elsewhere. Among the amendments prescribed in the Firearms Amendment (Trafficking) Act 2001 No 24 (NSW) was the stipulation that registration now ‘applies to every firearm frame and firearm receiver in the same way as it applies to a firearm’ (s 93(1)).

Long-arms, regardless of restricted status, were predominantly drawn from the grey market. Inconsistent inter-jurisdiction regulations on the registration of long-arms allowed a store of unregistered long-arms, including restricted models, to accrue well before the 1996 National Firearms Agreement. This grey market of firearms has thus served, and probably continues to serve, as a reliable and well-stocked resource for the illicit market.

The trafficking of illicit handguns has relied on alternative methods of transfer, influenced in part by the traditionally stricter controls on handgun ownership and use. The Queensland deactivation loophole almost certainly contributed to the trafficking of illicit handguns and while the NFTD data does not allow confirmation of this assumption, the significance of this category in the findings strongly suggests it played a substantial role. Other forms of diversion were apparently much less important, as was illicit domestic manufacture and illegal import. Theft, however, seems to have made a reliable contribution. It was the source for 50 percent of non-restricted handguns and 31 percent of restricted handguns. These results, however, must be interpreted with caution as the data on the source of illicit handguns was largely incomplete.

The contrasting role of deactivation in the diversion of restricted and non-restricted handguns probably reflects the efforts that would be taken to distribute a coveted item (ie a restricted form of semi-automatic pistol). Theft is a risky enterprise but often an opportunistic one too; deactivation (and other complex forms of diversion) is more likely to be used for firearms of greater value and/or models that are in demand.

Data limitations

It is probable that the great majority of restricted handguns, like long-arms, were already in the illicit market well before the respective gun buybacks, but it is not clear whether this past supply has produced a pool of illicit firearms large enough to address current (or future) levels of demand. Reports in recent years of large-scale trafficking operations, alongside smaller, single-order transactions, indicate that additional supplementing, through illegal import, domestic manufacture and theft, has been occurring. To what extent recycling, rather than replenishing, characterises any of these operations, however, has not been considered; nor is data available to explore this matter further.

The analysis presented above describes the general characteristics of the illicit firearms market in Australia and how it has been sustained but it also reveals how dependent analysis of this type is on the availability of accurate, comprehensive data. It was possible with the available data to describe the composition of the illicit market, and the firearm preferences of serious and organised crime groups, but the validity of other findings, particularly around points of diversion, was affected by a substantial number of unknown responses. The poor quality of the recorded data also prevented any substantive comment on the contribution of illegal importation in supplying the market, potentially concealed information on diversion pathways for long-arms (with the grey market; being the de facto source assigned to most unregistered firearms); and precluded reliable identification of last legal ownership of illicit firearms. These deficiencies were not a problem of the dataset itself but rather an illustration of the deficiencies in the documentation of key firearm transactions, an issue affecting the tracing of firearms that is discussed in the next section.