Is Burglary a Violent Crime? An Empirical Investigation of the Armed Career Criminal Act’s Classification of Burglary as a Violent Felony

Abstract

Traditionally considered a non-violent property offense, burglary is nonetheless classified as a violent crime under the federal Armed Career Criminal Act (ACCA). The ACCA, a three-strikes law that provides a mandatory minimum sentence of 15 years, is triggered when an offender, who has been previously convicted for a crime classified under the ACCA as either a “violent felony” or “serious drug offense,” is convicted at the federal level for any felony committed while in possession of a firearm. The present study investigated the ACCA’s classification of burglary as violent through analysis of National Crime Victimization Survey data for the period of 2009 to 2014. Results showed that burglary is overwhelmingly a non-violent offense. The national incidence of actual violence or threats of violence during a burglary was 7.9%. At most, 2.7% of burglaries involved actual acts of violence. Legislative reform of the ACCA classification to match the empirical description of burglary is discussed.

Keywords

sentencing policy violence research and policy

In the late 1970s, the United States rejected the idea of rehabilitation (Cullen, 2005) and got “tough on crime” in response to increasing crime rates and the politicization of crime and public safety during the 1960s (Tonry, 2004). This hardened public opinion about crime prevention in the 1970s, merged with the public’s growing fear of drug-related crime in the 1980s, and led to major legislative efforts to toughen crime control policies in the 1980s and 1990s (Musto, 1999; Tonry, 2011; Windlesham, 1998). Spurred by newly enacted mandatory minimum sentences, three-strikes laws, truth in sentencing, and life without parole laws, the prison population dramatically increased as offenders received harsher sentences, served longer sentences prior to release, or both (Tonry, 2004). More recently, tough on crime sentiments have waned as re-entry, treatment, and restorative justice programs have proliferated across the United States (Cullen, 2005; Tonry, 2011). However, many tough on crime era statutes remain in effect, with the Armed Career Criminal Act (ACCA) being just one example.

The ACCA

Passed as part of the Federal Comprehensive Crime Control Act of 1984 and recodified and amended in 1986, the ACCA (18 U.S.C. § 924, 1986) is a focused three-strikes law that provides a mandatory minimum sentence of 15 years to life in prison, in addition to a fine not to exceed US$25,000. The ACCA is triggered when an offender, who has been previously convicted at either the state or federal level for a crime classified under the ACCA as either a “violent felony” or “serious drug offense,” is convicted at the federal level for any felony crime committed while in possession of a firearm (a violation of 18 U.S.C. § 922). Although violence can be defined in emotional or psychological terms, the ACCA and the present study define violence as threatened, attempted, or inflicted physical harm or injury to the person of another.

What constitutes a violent felony under the ACCA (1986) is defined using a two-pronged definition. First, a violent felony is any crime that has as an element “the use, attempted use, or threatened use of physical force against the person of another” (18 U.S.C. § 924(e)(2)(B)(i)) or second, “is burglary, arson, extortion, involves use of explosives or otherwise involves conduct that presents a serious potential risk of physical injury to another” (18 U.S.C. § 924(e)(2)(B)(ii)), also referred to as the “otherwise provision.” The elements and other offenses used by the ACCA to discern violent felonies inherently involve either actual physical force or injury, or create situations highly likely to result in physical force or injury. Burglary, defined as entering or remaining in a structure with the intent to commit a crime therein (Culp, Kopp, & McCoy, 2015), has long been considered a non-violent property crime and is the outlier.

The House and Senate Reports detailing the debate of the act provide some insight into the act’s classification of burglary as a violent offense. Senator Arlen Specter (Republican from Pennsylvania; R-PA) viewed burglary as one of the “most damaging crimes to society,” involving the “invasion of [victim’s] homes and workplaces, violation of their privacy, and loss of their most personal and valued possessions” (H.R. Rep. No. 98-1073, 1984, p. 3). The Senate Report stated that burglary was one of “the most common violent street crimes” and that while it “is sometimes viewed as a non-violent crime, its character can change rapidly, depending on the fortuitous presence of the occupants of the home when the burglar enters, or their arrival while he is still on the premises” (S. Rep. No. 98-190, 1983, pp. 4-5). When asked in 1986, during the recodification and amendment debate, why burglary should be retained as a predicate offense for enhancing the severity of sentencing, Deputy Assistant Attorney General James Knapp responded, “though injury is not an element of the offense, it is a potentially very dangerous offense, because . . . there is a very serious danger to people who might be inadvertently found on the premises” (Armed Career Criminal Act: Hearing on H.R. 4639, 1986, p. 26).

In Taylor v. United States (1990), the United States Supreme Court upheld the ACCA’s classification of burglary. Taylor’s sentence for being a felon in possession of a firearm had been enhanced under the ACCA based upon a criminal history including two prior convictions in Missouri for second-degree burglary, an offense with no violent elements. Accordingly, Taylor argued that these convictions did not qualify as violent felonies under the ACCA because they did not involve violent conduct (Taylor v. United States, 1990). The Court held that Taylor’s three prior convictions were violent under the ACCA, finding that the broadest definition of burglary must be used because of wide variation in definitions among the individual states. In the Court’s opinion, by specifically stating in the statute that burglary was violent, Congress meant to regard all burglaries committed in the states as also violent and not simply refer to the subset of burglaries in which the offender’s actual violent conduct or possession of a weapon increased the risk of physical injury (Taylor v. United States, 1990).

More recently, the United States Supreme Court has examined if attempted burglary can be classified as a violent crime. The Court held in James v. United States (2007) that attempted burglary was also a violent crime. James had argued that his three prior convictions for attempted burglary did not qualify as violent crimes and should not have triggered an enhancement of his sentence under the ACCA. The Court found that while James’s previous Florida convictions for attempted burglary did not meet the generic definition of burglary explicated in Taylor, attempted burglary did satisfy the ACCA’s provision for crimes that “otherwise involve conduct that presents a serious potential risk of physical injury to another” (ACCA, 18 U.S.C.S. § 924(e)(2)(B)). The Court went on to state that attempted burglary, while sustaining the same risks as completed burglary, may indeed pose a risk even greater than that of a completed burglary as many completed burglaries do not involve confrontations, whereas attempted burglaries often do. However, most recently, in Johnson v. United States (2015), the Court struck down the ACCA’s otherwise provision cited in James, nullifying its classification of attempted burglary as violent.

The ACCA’s classification of burglary as violent raises two central questions. First, is burglary inherently violent? Second, does burglary pose a serious potential risk of physical injury? The present study addressed these questions by using 6 years of data from the National Crime Victimization Survey (NCVS) to estimate the incidence and nature of violence that occurs during burglaries.

Incidence of Violence During Burglary

Conklin and Bittner (1973) first explored the frequency of contact between burglars and residents and attendant levels of violent incidents in their analysis of 945 burglaries reported to the police in a Northeast United States suburb of 100,000. They found the majority (63.7%) of the burglaries occurred at a home or residence. Almost half (39%) occurred on the weekend between 7 p.m. to 10 p.m. when most residents were away. Most relevant to the current study, Conklin and Bittner (1973) found that across all types of burglary, only 2.5% involved contact between victim and offender.

The burglary-violence connection was further unpacked in the United Kingdom by Budd (1999), using data from the 1998 British Crime Survey. Findings revealed that victims reported “violent or threatening behavior” occurred during 11% of residential burglaries. Unfortunately, “violent” was not disaggregated from the “threatening” aspects of burglar behavior, precluding the determination of the actual extent of physical violence that occurred during burglaries.

Rand (1985) explored the burglary-violence connection using data from the National Crime Survey (the predecessor of the NCVS), which included roughly 73 million incidents of household burglary (Rand, 1985). The most noteworthy finding was that only 3.8% of household burglaries involved violence. Notwithstanding the low level of overall violence and forced entry, the 3.8% violence rate amounted to 2.8 million incidents over the 10-year study period, of which 39% of these total incidents (i.e., 39% of the 3.8% of cases that involved violence) were simple assaults, 23% were aggravated assaults, 28% were robberies, and 10% were rapes. Rand’s (1985) findings were cited in the influential Supreme Court case of Tennessee v. Garner (1985), in which the Court ruled that it was unconstitutional for police to use deadly force in trying to stop an unarmed, non-dangerous, fleeing burglary suspect. The Court cited the low incidence of violence in crimes of burglary as evidence that the fleeing felon in this case was non-violent.

Rand (1985) also disaggregated types of burglary (e.g., forcible entry, attempted forcible entry, and unlawful entry) and various household types (owned or rented, family income level, race of residents, urban/suburban/rural location, etc.). Findings indicated that nearly 40% of household burglaries were committed by someone related to or known by the victims, and that a theft occurred in 60% of incidents. Rand (1985) noted that only 12.7% of incidents occurred while a person was home, suggesting that most burglars seek to avoid contact with their potential victims. Rand (1985) further noted that burglaries in which someone was home or violence had occurred are reported to police more often than either victim-absent or non-violent burglaries, suggesting that the true percentage of cases in which burglaries involve no violence may be even higher than statistics record. Put another way, the percentages might be overestimates of the occurrence of violent burglary. Rand’s (1985) study provided the inspiration and the model for the current research being reported here.

More recently, two independent studies of burglary and violence reached similar conclusions. Catalano (2010), who used NCVS data from 2003 to 2007 to update the work of Rand (1985), found that an estimated 3.7 million burglaries occurred during each year of the study’s period. Compared with Rand (1985), Catalano (2010) found a higher rate of violence for burglaries, with violence or threats of violence occurring during 7.2% of all burglaries. In 27.6% of these incidents, a household member was present, and 65.1% were committed by someone known to the victims. Similarly, Culp et al. (2015) used NCVS, Uniform Crime Report (UCR), and National Incident Based Reporting System (NIBRS) data for the period from 1998 to 2007 to explore the incidence of violence that occurred during a burglary. Most notably, they concluded the incidence of violence that occurred during burglaries varied from as high as 7.6% in urban cities to a low of 0.9% in rural areas of the United States. In addition, the incidence of actual violence (as indicated by physical injury to a victim) was only 2.7%, while 4.9% of offenders threatened violence or harm but there were no physical injuries. Finally, a victim was present (although not necessarily cognizant of the offender’s presence) during 26.0% of all burglaries recorded in the NCVS. Although the data sources are not necessarily exact matches, the percent of burglaries that involved violence rose to a high of between 7.2% (Catalano, 2010) and 7.6% (Culp et al., 2015) from 3.8% (Rand, 1985), suggesting burglaries may be increasingly likely to involve some form of violence.

Burglar Target Selection

Ethnographic research based on offender self-reports has suggested that burglars offend for profit, select their targets with care, plan their crimes, and take into account multiple variables in deciding when and where to offend (Bennet & Wright, 1984; Cromwell, Olson, & Avary, 1991; Nee & Taylor, 2000; Wright & Decker, 1994). Burglars perform a criminal calculus in line with the rational choice theory of crime, weighing the costs and benefits of their illicit acts before taking action (Cornish & Clarke, 1986). However, Cromwell et al. (1991) posited that “a completely rational model of decision making in residential burglary cannot be supported” (p. 43). Johnson and Bowers (2004), borrowing from behavioral ecology, described burglars as “optimal foragers” who, like scavengers, seek “to increase the rate of reward while minimizing both the amount of time searching for food and the risk of being attacked” (p. 242). Bernasco and Nieuwbeerta (2005) added that burglars seek to maximize rewards by selecting targets “that require little effort to enter, that appear to contain valued items, and that give the impression that the likelihood of being disturbed or apprehended there is low” (p. 297). Indeed, several studies have highlighted the decision process of burglars (Bernasco & Nieuwbeerta, 2005; Townsley et al., 2015).

Accordingly, a burglar’s target selections are most likely, to varying degrees depending on the individual offender, the result of both rational choice and impulse. Offenders seek targets that appear to offer them the most profit (Bennet & Wright, 1984; Reppetto, 1974), while presenting the least risk of potential witnesses being present (Bennet & Wright, 1984; Cromwell et al., 1991). Ideal targets are familiar to the offender because they fit the offender’s knowledge of the community and the daily routines of its inhabitants (Brantingham & Brantingham, 1981). Offenders frequently live in or near the area (Bernasco & Luykx, 2003; Rengert & Wasilchick, 2000) or select it because they researched the area after identifying suitable targets. Wright and Decker (1994) found their sample of active offenders to have spent considerable time (days and weeks) learning the geographic layout and temporal routines of their intended targets.

Finally, given the effort exerted in selecting and learning a target area, offenders seek to maximize their profits by concentrating on areas that contain many potential targets. Several studies have found that once a residence has been burglarized, its risk of being burglarized again increases for a short period, repeat victimization (Budd, 1999; Hearnden & Magill, 2004), as does the risk that other residences in the neighborhood will be burglarized, near-repeat victimization (Bernasco, 2008; Bowers & Johnson, 2005; Johnson et al., 2007).

Summary

In contrast with the ACCA’s classification of burglary as a categorically violent offense, the review of research on the incidence of violence that occurs during burglaries found that a small percentage of burglary offenses result in violence. Although research into burglars’ target selection found that burglary offenders go to great lengths to avoid their human victims, offender’s avoidance of victims may explain why so few burglaries have been violent. This contradiction between policy and empirical findings raises the central question taken up by the present study. Is burglary violent?

Method

Data

The present study used the NCVS for the period from 2009 to 2014 to estimate the incidence of violence that occurs during a burglary. The NCVS contains a representative sample of United States households, in which occupants were surveyed regarding any criminal victimization they may have experienced in the previous 6 months. The NCVS gathers detailed information regarding the crimes of rape/sexual assault, robbery, aggravated and simple assault, purse-snatching/pocket-picking, burglary, theft, and motor vehicle theft, as well as demographic information that facilitates the weighting of respondents’ answers to estimate national victimization trends. Based upon this information, the NCVS determine what, if any, types of victimization occurred utilizing a hierarchical classification system, as well as weighting variables based upon the demographics of the respondents that allow for the estimation of national victimization rates. The NCVS, however, does not collect information of the crime of murder, a limitation to the present analysis. While, undoubtedly, some homicides co-occur during a burglary, homicide is the least frequent of index offenses (Federal Bureau of Investigation, U.S. Department of Justice, 2016).

The NCVS represents the only nationally representative data collection in the United States that facilitates the identification and exploration of co-occurring criminal offenses. Through the Inter-university Consortium for Political and Social Research, the Bureau of Justice Statistics annually makes available an NCVS data file at the address, household, respondent, and incident levels of measurement. Crime incidents served as the units of analysis for the present research.

Measures

Identifying violent burglaries

The hierarchical classification of offenses in national datasets is the biggest roadblock to the estimation of the extent of violent burglary. In a hierarchical system, offenses are classified as the most serious or severe crime that occurred during an incident. However, it is common for crimes, such as burglary, to co-occur with other more severe crimes. In a hierarchical coding system, a burglary incident that co-occurred with a more serious violent offense will not be counted in favor of reporting the more serious offense. To identify the incidence of violence that occurs during burglaries, first, the hierarchal classification of offenses must be untangled. To do this, one must identify burglary incidents hidden by the more serious co-occurring offenses. Doing so allowed the present study to report how many burglaries were violent, at what level of violence, and whether household members were present at the time of the burglary.

The present study solves the hierarchy problem by using information already contained in the NCVS. The NCVS asks respondents a series of questions about what kind, if any, victimization they had experienced in the last 6 months. Using that information, the NCVS classifies incidents by type of crime and report detailed information for that victimization. Burglary is defined as entering or remaining in a structure with the intent to commit a crime therein (Culp et al., 2015). Using this definition, an offender has committed burglary when they enter a place with or without force, and do not have the right to be there. Based upon these required elements, three questions in the NCVS are central to the classification of an incident as a burglary: (1) “did offender have a right to be there,” (2) “did offender get inside,” and (3) “evidence of forcible entry.” The present study recoded the responses to these questions and used them to create an additive scale variable that identifies any incident in which a burglary occurred.

This variable is then combined with the NCVS’s type of crime variable to create the additive variable utilized here to identify co-occurring offenses.¹ For example, if the variable identifying burglary incidents is coded as burglary (20), non-burglary (0), and the variable identifying type of crime is coded rape/sexual assault (200), robbery (300), aggravated assault (400), simple assault (500), and burglary (600). When these codes are added together to create a new type of crime variable, an incident coded 220 would be rape/sexual assault that co-occurred with a burglary, while an incident coded 200 would be a rape sexual/assault that did not involve burglary.

Other measures

For violence to occur, the victim and offender must converge in space and time. The measure—household member present—was a dichotomous variable that indicated if the respondent or a member of their household was present at any time during the offense. All NCVS respondents were asked this question for all crimes.

The type of violence that occurred was measured by two different variables. The first—threatened/attacked—was a categorical variable that indicated if the victim was not threatened or attacked, threatened but not physically attacked, or physically attacked during the incident. To create the variable, the three dichotomous responses were combined into a single measure.

The second indicator was a measure of physical injury, which identified the nature and severity of injury suffered by the victim using the following categories: no injury, cuts/bruises, broken bones, rape/sexual injury, unconscious with broken bones, penetration wounds (stabbing or puncture), penetration wounds with rape/sexual injury, internal injury, internal injury with rape/sexual injury, and other injury. The variable was created by combining the 11 dichotomous variables into a single measure. Any NCVS respondents who reported being physically attacked during an offense were asked these questions.

Analysis

The present study’s central purpose was to identify the incidence of violence that occurs during the crime of burglary, in addition to indicators of the type and severity of violence that occurred to inform sentencing policy decisions. Frequency distributions describing the incidence of co-occurrence between burglary and violent crime, how often a household member was present during a burglary, if a victim was attacked and the type of attack, and the type of injuries suffered, if any are reported, to support the study’s findings and recommendations.

Results

Between 2009 and 2014, the NCVS estimated that 18,989,132 burglaries occurred in the United States. The goal of the present study was to identify additional burglaries that may have been “hidden” because they co-occurred with a more severe violent offense. During the study period, an additional 1,637,454 incidents involving the co-occurrence of burglary with a violent crime (rape/sexual assault, robbery, aggravated assault, and simple assault) were identified. The addition of these incidents increases the total number of burglaries estimated by the NCVS to have occurred during the study period to 20,626,586, representing an 8.60% increase in burglary offenses.

Based on the findings presented in Table 1, these violent burglaries comprised 7.9% of all burglaries estimated to have occurred between 2009 and 2014 in the United States, an estimate close to, and slightly higher than the figures produced by Culp et al. (2015) and Catalano (2010). Simple assault was the violent offense to most likely co-occur with burglary (4.2%), followed by robbery (1.7% of burglaries co-occurred with robbery). As depicted in Figure 1, the overall incidence of violence remained relativity stable over the study period, varying between 7.5% and 9.3%. The sole exception was a dramatic decrease to 5.9% in 2013 followed by an equally dramatic increase to 9.3% in 2014. To further explore this trend, Figure 2 presents the incidence trends of violent crime (rape/sexual assault, robbery, aggravated assault, simple assault), non-violent burglary, and violent burglary. Overall, the incidence of violent burglary remained stable, only fluctuating by roughly 57,000 incidents, with the exception of a dramatic decrease of 101,000 incidents in 2013, followed by a dramatic return to just short of 2012 occurrences in 2014.

Table 1.

Estimate of the Co-Occurrence of Burglary and Violence in the NCVS: 2009-2014.

Offense type	Total offenses	Co-occurred with burglary	% within offense	% of burglaries that involved violence
Rape/sexual assault	1,270,047	113,205	8.9	0.5
Robbery	3,285,811	352,524	10.7	1.7
Aggravated assault	5,065,566	305,119	6.0	1.5
Simple assault	17,646,255	866,606	4.9	4.2
Subtotal		1,637,454		7.9
Burglary	18,989,132	18,989,132	100	0
Total	46,256,811	20,626,586		7.9

Note. NCVS = National Crime Victimization Survey.

Figure 1.

Percentage of co-occurrence of burglary and violence in the NCVS by year; 2009-2014.

Figure 2.

Incidence of violent crime, burglary, and violent burglary in NCVS by year; 2009-2014.

As would be expected, overall trends in both violent crime and burglary influence trends in the co-occurrence of the two offenses. In 2013, both violent crime and burglary decreased (Truman & Langton, 2014), and, therefore, the co-occurrence of these offenses also decreased. The following year, the incidence of violent crime showed no significant change, while the incidence of burglary decreased significantly (Truman & Langton, 2015). These trends would explain the dramatic increase in the percentage of violent burglaries depicted in Figure 1 for 2014. However, the dramatic increase in the estimated incidence of violent burglaries in 2013, while overall burglary declined and violent crime did not change, defies explanation and warrants future study.

The presence of a potential victim is a necessary condition for a violent burglary to occur. Stated more simply, for burglary to be violent, a victim must be present for violence to occur or be threatened. The review of previous burglar behavior research indicated that burglars go to great lengths to avoid occupied targets (Wright & Decker, 1994). Supplemental analyses of the NCVS data support this, finding that a household member was home or came home in about 27.5% of all burglaries between 2009 and 2014. Expressed differently, roughly 72% of all burglaries between 2009 and 2014 in the United States precluded the occurrence of violence.

Thus far, all types of violence have been aggregated under the general label of violence. Included in violent burglaries were incidents where a victim was threatened with harm but not physically attacked, as well as physical attacks, whether attempted and completed. Table 2 disaggregates victim-present burglaries into incidents of no violence, threats, and actual physical attacks. In 68.1% of incidents, a victim was neither threatened nor attacked. In the minority of burglaries that resulted in violence, 14.5% of victims were threatened but not physically attacked, and 17.4% (attempted and completed attacks combined) resulted in some degree of physical attack upon the victim. Incidents during which a victim was physically attacked represented 4.35% of all burglaries that occurred during the 6-year study period.

Table 2.

Type of Violence That Occurred During Burglaries in NCVS: 2009-2014.

	Incidence	% of victim present	% of total burglaries
Not threatened nor attacked	3,522,006	68.1	92.0
Threatened	751,340	14.5	3.6
Attempted attack	165,794	3.2	0.8
Attacked	730,751	14.1	3.5
Total^a	5,169,891	100.0	100.0

Question was only asked if victim was present during offense. NCVS = National Crime Victimization Survey.

While 70.8% of those attacked reported injury, about half of victims (50.8%) reported only cuts and bruises. Table 3 presents the type of injuries reported by victims of violent burglaries. After cuts and bruises, the most prevalent injuries reported were injuries consistent with rape/sexual assault (5.5%), broken bones or teeth (4.4%), penetration wounds (2.2%), and other types of injuries (4.9%). Incidents that resulted in an injury to a victim constitute 10.00% of all victim-present incidents, and just 2.50% of all burglaries that occurred during the study period.

Table 3.

Type of Injuries Suffered by Burglary Victims in the NCVS: 2009-2014.

	Incidence	% victims attacked
No injury suffered	213,375	29.2
Cuts/bruises	371,784	50.8
Broken bones	32,117	4.4
Rape/sexual assault injury	40,147	5.5
Unconsciousness with broken bones	2,403	0.3
Penetration wounds	16,250	2.2
Penetration wounds, and rape/sexual assault injury	6,463	0.9
Internal injury	7,184	0.9
Internal injury, and rape/sexual assault injury	4,836	0.6
Other	36,193	4.9
Total^a	730,752	100.0

Question only asked if victim suffered a completed attack. NCVS = National Crime Victimization Survey.

Discussion

Overall, the findings from the current analyses suggest that except in a small minority of cases, burglary is not a crime of violence. The majority of burglaries (about 73%) occur when household members are not present. As indicated by prior research, burglars go to great lengths to learn the patterns of their victims, ostensibly to avoid them (Bennet & Wright, 1984; Wright & Decker, 1994). This study, as well as past empirical research (Catalano, 2010; Culp et al., 2015; Rand, 1985), supports these findings. When violence is disaggregated by type (threat vs. attack), the argument that burglary is physically violent, as suggested by the ACCA, becomes even more tenuous. While a small minority of burglary victims do experience a physical attack (4.3%), an even smaller minority reports physical injuries. Victims reported being injured in 2.5% of all burglaries, with half of those who were injured reporting only cuts and bruises.

To say that burglary is seldom a crime of violence is not to denigrate the experience of burglary victimization. The psychological aftermath of burglary victimization can be as serious, if not more serious, than any physical trauma, and often goes unseen and untreated. Burglary victims report feelings of violation because their personal space has been intruded upon and their possessions rummaged through and taken, while also feeling vulnerable and potentially unable to protect themselves from future offenses (McCann, Sakheim, & Abrahamson, 1988). Victimization may also lead to elevated and perhaps disproportionate fear of crime (Skogan, 1987), and distrust of others, leading to increased levels of fear and anxiety, which can have debilitating and other physiological manifestations if not addressed (Janoff-Bulman & Frieze, 1983). Perhaps it is the emotional and psychological trauma that can result from burglary victimization that prompted legislator’s inclusion of burglary under the ACCA.

The law, however, can take account of the victim’s perception of violation without regarding the offense as violent. If the offender is apprehended, victims’ experiences are often considered during the sentencing of offenders in the form of victim impact statements (Lens, Pemberton, & Bogaerts, 2013; Myers & Greene, 2004). Alternately, the approach recommended by restorative justice is for victim advocates to “put a face” on the offender for the victim. That is, victims are often reassured when they learn who the offender is, why the offender committed the crime, why the offender chose that target, and to know that the offender will not be able to return (Angel et al., 2014; Sherman et al., 2005). Considering that in approximately 40% of all burglaries, the victim was personally acquainted with the offender in some way, this approach may be more helpful to victims than simply assuring them that punishment will be severe.

Although many burglary victims may feel violated, they very seldom will be physically harmed. However, under the ACCA, all completed burglaries are classified as violent felonies and counted as predicate offenses under a “three strikes”-type sentencing enhancement. Counting burglary as a violent crime rather than a property crime over-criminalizes the offense, and leads to considerably longer sentences for offenders whose sentences are enhanced under the ACCA.

Co-Occurrence of Burglary and Other Offenses

In criminal codes outlining illegal conduct, offenses are defined in terms of the illegal act committed and the intent to commit the illegal act. An assailant commits an assault with the intent to commit an assault. By definition, burglary is unique. A burglar does not enter or remain in a structure with the intent to enter the structure, his intent is to commit another, separate offense. Because of this, burglary often, if not always, co-occurs with other offenses.

Once an offender enters with intent to commit a crime, they are guilty of burglary regardless of whether or not the intended crime is completed. Most often, burglary co-occurs with theft. Theft, being a less severe property crime, merges with the burglary, and the offender is typically only charged with the more severe offense of burglary. But the offenses committed once inside can also be violent, and the nature and extent of the offender’s conduct determines what additional charges can be levied.

When an offender enters a structure and assaults a victim but does not commit theft, they are charged with burglary, for entering with the intent to commit a crime, and an additional offense for the assault. The specific violent offense the offender is charged with depends upon the nature and extent of the crime. In addition to the assault based charges, the majority of states elevate burglary from a simple to an aggravated offense when a victim is injured, threatened, or otherwise placed in fear (Culp et al., 2015).

To remove burglary from the ACCA’s list of categorically violent offenses would not result in violent offenders going unpunished. The small subset of burglary cases in which violence actually occurred or was threatened are important and, if punished as violent crimes, would not over-criminalize the offense because they would be regarded as what they are: rapes, robberies, and assaults in addition to burglaries. All of these violent burglaries, or burglaries that co-occurred with a violent offense, would continue to be covered by the ACCA because they involve “the use, attempted use, or threatened use of physical force against the person of another or involves use of explosives” (ACCA, 18 U.S.C. § 924(e)(2)(B)(i)).

What the ACCA Costs

The ACCA’s classification or seeming misclassification of all burglaries as violent exacts a cost not only on the offender whose prison sentence is enhanced, but also on the criminal justice system for imposing it. Severe punishments should be reserved for serious offenses. Beccaria (1775/1983) noted that “thefts without violence should be punished with fines . . . but when violence is added to theft, then punishment ought to be likewise a mixture of corporal punishment and penal servitude” (p. 53). Holmes (1881/2004), addressing all justifications for punishment, argued that “punishment must be equal, in the sense of proportionate to the crime” (p. 31). Beccaria (1775/1983) and Bentham (1823), based on common sense ideas of equity, justice, and fairness, as well as utilitarian concerns, argued that the severity of punishment should correspond to the severity of the offense. Von Hirsch (1976) argued in his “just deserts” philosophy of punishment that the misclassification of an offense leads to the miscalculation of the offense’s severity, which then leads to the levying of a sentence disproportionate to the offense’s actual severity.

According to annual United States Sentencing Commission (USSC) data from 2009 to 2014, a total of 3,664 offenders were sentenced under the ACCA. The specific crimes constituting the criminal histories that qualified these offenders for sentencing enhancements are not recorded in the published statistics of the USSC and can only be determined through an analysis of USSC master data files, including Pre-Sentence Reports. However, utilizing State Court Processing Statistics from the Bureau of Justice Statistics (Reaves, 2013) for the period 1990 to 2009, a rough estimate of the number of offenders convicted under the ACCA enhancement can be derived. Because the ACCA counts all burglaries as violent crimes, it follows that the proportion of offenders convicted under the ACCA enhancement would approximate the proportion of offenders convicted for felony burglary. Between 1990 and 2009, burglary comprised 7.2% of all felony convictions in the 75 largest counties in the United States, according to the ACCA.

Applying this percentage to the total number of offenders sentenced under ACCA enhancements, it is estimated that in the 6-year period from 2009 to 2014, approximately 264 offenders received sentencing enhancements due to a non-violent burglary being considered a crime of violence. Federal defendants who have had a non-violent burglary counted as a violent offense and are subsequently sentenced as an Armed Career Criminal receive a sentence enhancement of 5 years. This represents the difference in the mandatory minimum sentence proscribed for violation of 18 USC 922(g) (10 years; United States Sentencing Commission, 2016a, § 2K2.1.)—felon in possession of a firearm—and violation of 18 USC 922(g) with the ACCA enhancement 18 USC 924(e) (15 years; United States Sentencing Commission, 2015, 2016c, § 4B1.4). Over 5 years, it is estimated that the 264 inmates who have been sentenced under this practice are serving a collective total of 1,320 additional years in prison.

Beyond the human toll exacted on the inmates who are affected by it, the levying of disproportionate sentences under the ACCA can also affect the publics’ general compliance with the law. Tyler (2006) argued that when individuals regard the legal system as legitimate, they are more likely to comply with the law. Individuals, based on their ideas of justice and fairness, desire fair procedures that result in fair outcomes. To establish and maintain legitimacy, the legal system must provide outcomes consistent with the theories of distributive and procedural justice. Distributive justice deals with the outcome of a case, suggesting that people desire to receive the level of punishment they feel they deserve. Procedural justice, rather, focuses on how the outcome of a case was reached: “if the judge treats them fairly by listening to their arguments and considering them, by being neutral, and by stating good reasons for his or her decision” (p. 6). Failure to provide fair outcomes affects the legitimacy with which the system is regarded, which in turn affects the degree to which individuals obey the law (Anderson, 1978; Sherman, 1993; Tyler, 2006). Disproportionate sentencing violates the theory of distributive justice and society’s and individuals’ innate sense of justice and fairness, and may result in loss of legitimacy.

Moving past moral and practical arguments, the ACCA’s sentencing enhancement is expensive, and depletes public funds that might otherwise be used in other areas of criminal justice. According to the Department of Justice, the average annual cost (in 2014 dollars) per inmate of the federal prison system amounts to US$30,619 (Federal Bureau of Prisons, Annual Determination of Average Cost of Incarceration, 2015). If we multiply the annual cost by the aggregate years added by the non-violent burglary enhancement, this gives us a total estimated cost of the practice of US$40,417,080 over 6 years, or an average of US$6.7 million per year.

Policy Recommendation

Reform of the federal ACCA might prove to be quite simple. All burglaries were included and classified as violent felonies under the Act, as a result of the United States Supreme Court interpreting Congressional intent in passing the Act. A very simple amendment would make a big difference: removing the word “burglary” from 18 U.S.C § 924(e)(2)(B)(ii) of the act could definitively signal that the Congressional intent was not to regard burglary as violent. This basic alteration would remove non-violent offenders from the reach of the ACCA, while the statute’s enhancements would continue to be applied to violent burglars through subsection (i) of the Act. Of course, this would not preclude judges from adding additional sentencing enhancement “points” for each proven factor involving violence (e.g., presence or use of a weapon, vulnerability of the victim, threats) as is regularly the practice in federal sentencing (United States Sentencing Commission, 2016d). These additional points would raise the seriousness level proportionate to what actually occurred.

In August 2015, the USSC, in response to the United States Supreme Court’s ruling in Johnson v. United States (2015), amended the Career Offender provisions of the United States Sentencing Guidelines. Previously, the guidelines had classified all burglaries of a dwelling, whether attempted or completed, as crimes of violence. Proposed amendments circulated by the Commission for comment would have expanded this classification to all burglaries (United States Sentencing Commission, 2015). However, as a result of public comments and review of available research, the Commission, instead, removed burglary from the provisions list of included offenses (United States Sentencing Commission, 2016b, § 4B1.1).

Limitations and Future Research

Nevertheless, there were limitations to the current study. The study’s first limitation stems from the datasets used to derive the estimates of violence. The NCVS, while widely used by social science to identify and analyze national criminological phenomena, is based upon survey information. Respondents’ answers are weighted based upon demographic factors to produce estimated offense totals. For this reason, NCVS results are often validated against offense totals from the UCR. Because the present study’s interest was in the co-occurrence of offenses that cannot be identified in UCR data, it was impossible to perform this validation.

The study’s second limitation stems from the lack of data on the emotional impact of burglary. As previously discussed, violence can be both physical and emotional. The current study provides description of the physical aspects of burglary, but was unable to also describe the emotional aspects of burglary victimization. Ideally, both the physical and emotional impacts of burglary would be assessed at the same time to provide the most complete description of the offense. Future research should look at the emotional aspects of burglary victimization. The NCVS began querying respondents about the psychological impacts of the victimization they suffered in 2009. However, during the present study period, only respondents who were physically attacked are asked these questions, and not those who suffered only property loss.

Future research should also investigate the mechanisms behind the increase in the co-occurrence of violence and burglary that occurred during the period from 1982 to 2007. The increase in violent burglary during a period when all other types of crime were declining provokes interest. One potential explanation worthy of inquiry is that more offenders were forced to target their victims in their residences instead of the street because of changes in monetary systems (inception and proliferation of credit and debit cards).

Conclusion

The findings reported here demonstrate how important it is to differentiate between property crimes and violent crimes not only to prevent over-criminalization but also to address public and victims’ concerns carefully and fairly. By regarding burglary as a crime against property unless violence was actually involved, these concerns are met. By not including burglary in the group of offenses in which sentencing enhancements for violent offenses will attach—unless actual violence or weapons-carrying, indeed, occurred in addition to the burglary—over-punishment is also avoided. The policy implications of reforming the laws of burglary are powerful: public concerns to punish violent offenders would still be addressed, victims frightened in the sanctity of their homes will receive special attention, but offenders who have committed crimes against property will not serve long, expensive prison terms.

Footnotes

Author’s Note

This manuscript updates findings from the author’s doctoral dissertation completed at the City University of New York. The author would like to thank Candace McCoy and Brittany Hayes their review of earlier versions of this manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: While the author’s previous research on the incidence of violence that occurs during burglary was supported by a grant from the National Institute of Justice. The updated findings reported here were not supported by any agency or organization.

Notes

Author Biography

Phillip M. Kopp is an assistant professor of Criminal Justice in the Division of Politics, Administration, and Justice at California State University, Fullerton. In addition to his work on the incidence of violence in the crime of burglary, he is also working with police agencies to evaluate the use body-worn camera initiatives, and a use-of-force program that reengineers how force is used by police.

References

Anderson

(1978). A Place on the Corner. Chicago: University of Chicago Press.

Angel

Sherman

Strang

Ariel

Inkpen

Keane

Richmond

(2014). Short-term effects of restorative justice conferences on post-traumatic stress symptoms among robbery and burglary victims: A randomized controlled trial. Journal of Experimental Criminology, 10, 291-307.

Armed Career Criminal Act: Hearing on H.R. 4639 and H.R. 4768 before the Subcommittee on Crime of the House Committee on the Judiciary, 99th Cong, 2d Sess. (1986).

Armed Career Criminal Act of 1986, 18 U.S.C. § 924. (1986).

Beccaria

(1983). An essay on crimes and punishment (4th ed.). Boston, MA: International Pocket Library. (Original work published 1775)

Bennet

Wright

(1984). Burglars on burglary: Prevention and the offender. Aldershot, England: Gower Publishing.

Bentham

(1823). An introduction to the principles of morals and legislation. London, England: Printed for W. Pickering.

Bernasco

(2008). Them again? Same-offender involvement in repeat and near repeat burglaries. European Journal of Criminology, 5, 411-431.

Bernasco

Luykx

(2003). Effects of attractiveness, opportunity and accessibility to burglars on residential burglary rates of urban neighborhoods. Criminology, 41, 981-1001.

10.

Bernasco

Nieuwbeerta

(2005). How do residential burglars select target areas? British Journal of Criminology, 44, 296-315.

11.

Bowers

Johnson

(2005). Domestic burglary repeats and space-time clusters: The dimensions of risk. European Journal of Criminology, 2, 67-92.

12.

Brantingham

(1981). Notes on the geometry of crime. In Brantingham

Brantingham

(Eds.), Environmental criminology (pp. 27-54). Prospect Heights, IL: Waveland Press.

13.

Budd

(1999). Burglary of domestic dwellings: Findings from the British crime survey (Home Office Statistical Bulletin Issue 4/99). London, England: Home Office.

14.

Catalano

(2010). Bureau of Justice Statistics. Special report: Victimization during household burglary. Rockville, MD: Bureau of Justice Statistics.

15.

Conklin

Bittner

(1973). Burglary in a suburb. Criminology, 11, 206-232.

16.

Cornish

Clarke

(1986). The reasoning criminal: Rational choice perspectives on offending. New York, NY: Springer.

17.

Cromwell

Olson

Avary

(1991). Breaking and entering. Sherman Oaks, CA: Sage.

18.

Cullen

(2005). The twelve people who saved rehabilitation: How the science of criminology made a difference. Criminology, 43, 1-42.

19.

Culp

Kopp

McCoy

(2015). Is burglary a crime of violence? An analysis of national data 1998-2007 (Report No. 248651). Washington, DC: National Institute of Justice.

20.

Federal Bureau of Investigation, U.S. Department of Justice. (2016, September). Crime in the United States, 2015. Retrieved from https://ucr.fbi.gov/crime-in-the-u.s/2015/crime-in-the-u.s.-2015/offenses-known-to-law-enforcement/offenses-known-to-law-enforcement

21.

Federal Bureau of Prisons, Annual Determination of Average Cost of Incarceration 80 Fed. Reg. 12523. (2015).

22.

Hearnden

Magill

(2004). Decision-making by house burglars: Offenders’ perspective (Home Office Research Findings 249). London, England: Home Office.

23.

Holmes

(2004). The common law. New York, NY: Barnes & Noble. (Original work published 1881)

24.

H.R. Rep. No. 98-1073, 98th Cong. (1984).

25.

James, v. United States, 550 U.S. 192. (2007).

26.

Janoff-Bulman

Frieze

(1983). A theoretical perspective for understanding reactions to victimization. Journal of Social Issues, 39, 1-17.

27.

Johnson

Bernasco

Bowers

Elffers

Ratcliffe

Rengert

Townsley

(2007). Space-time patterns of risk: A cross national assessment of residential burglary victimization. Journal of Quantitative Criminology, 23, 201-219.

28.

Johnson

Bowers

(2004). The stability of space-time clusters of burglary. British Journal of Criminology, 44, 55-65.

29.

Johnson, v. United States, 135 S. Ct. 2551. (2015).

30.

Lens

Pemberton

Bogaerts

(2013). Heterogeneity in victim participation: A new perspective on delivering a victim impact statement. European Journal of Criminology, 10, 479-495.

31.

McCann

Sakheim

Abrahamson

(1988). Trauma and victimization: A model of psychological adaptation. The Counseling Psychologist, 16, 531-594.

32.

Musto

(1999). The American disease: Origins of narcotic control (3rd ed.). New York, NY: Oxford University Press.

33.

Myers

Greene

(2004). The prejudicial nature of victim impact statements: Implications for capital sentencing policy. Psychology, Public Policy, and Law, 10, 492-515.

34.

Nee

Taylor

(2000). Examining burglars’ target selection: Interview, experiment or ethnomethodology? Psychology Crime & Law, 6, 45-59.

35.

Rand

(1985). Bureau of Justice Statistics. Bulletin: Household burglary. Rockville, MD: Bureau of Justice Statistics.

36.

Reaves

(2013). Felony defendants in large urban counties, 2009—Statistical tables. Rockville, MD: Bureau of Justice Statistics.

37.

Rengert

Wasilchick

(2000). Suburban burglary: A tale of two suburbs. Springfield, IL: Charles Thomas.

38.

Reppetto

(1974). Residential crime. Cambridge, MA: Ballinger.

39.

S. Rep. No. 98-190, 98th Cong. (1983).

40.

Sherman

(1993). Defiance, deterrence, and irrelevance: A theory of the criminal sanction Journal of Research in Crime and Delinquency, 30, 445-473.

41.

Sherman

Strang

Angel

Woods

Barnes

Bennett

Inkpen

(2005). Effects of face-to-face restorative justice on victims of crime in four randomized, controlled trials. Journal of Experimental Criminology, 1, 367-395.

42.

Skogan

(1987). The impact of victimization on fear. Crime & Delinquency, 33, 135-154.

43.

Taylor v. United States, 495 U.S. 575. (1990).

44.

Tennessee v. Garner et al., 471 U.S. 1. (1985).

45.

Tonry

(2004). Thinking about crime: Sense and sensibility in American penal culture. New York, NY: Oxford University Press.

46.

Tonry

(2011). Retributivism has a past: Has it a future? New York, NY: Oxford University Press.

47.

Townsley

Birks

Bernasco

Ruiter

Johnson

Baum

White

(2015). Burglar target selection: A cross-national comparison. Journal of Research in Crime & Delinquency, 52, 3-31.

48.

Truman

Langton

(2014). Bureau of Justice Statistics. Bulletin: Criminal victimization, 2013. Rockville, MD: Bureau of Justice Statistics.

49.

Tyler

(2006). Why People Obey the Law. Princeton, NJ: Princeton University Press.

50.

Truman

Langton

(2015). Bureau of Justice Statistics. Bulletin: Criminal victimization, 2014. Rockville, MD: Bureau of Justice Statistics.

51.

United States Sentencing Commission. (2016a). Guidelines Manual. §2K2.1. Unlawful Receipt, Possession, or Transportation of Firearms or Ammunition; Prohibited Transactions Involving Firearms or Ammunition. Retrieved from http://www.ussc.gov/Guidelines/2016_Guidelines/index.cfm

52.

United States Sentencing Commission. (2016b). Guidelines Manual. §4B1.1. Career Offender. Retrieved from http://www.ussc.gov/Guidelines/2016_Guidelines/index.cfm

53.

United States Sentencing Commission. (2016c). Guidelines Manual. §4B1.4. Armed Career Criminal. Retrieved from http://www.ussc.gov/Guidelines/2016_Guidelines/index.cfm

54.

United States Sentencing Commission. (2016d). Guidelines Manual. § 5K. Departures. Retrieved from http://www.ussc.gov/Guidelines/2016_Guidelines/index.cfm

55.

United States Sentencing Commission, Proposed Amendment to the Sentencing Guidelines, Crime of Violence, 80 Fed. Reg. 70294. ( 2015).

56.

von Hirsch

. (1976). Doing justice: The choice of punishments. Toronto, Ontario, Canada: McGraw-Hill Ryerson.

57.

Windlesham

(1998). Politics, punishment, and populism. New York, NY: Oxford University Press.

58.

Wright

Decker

(1994). Burglars on the job: Streetlife and residential break-ins. Boston, MA: Northeastern University Press.