U.S. Department of Justice Office of Justice Programs Bureau of Justice Statistics This file was revised on 5/20/99 to be consistent with printed report. Special Report Carjackings in the United States, 1992-96 March 1999, NCJ 171145 By Patsy Klaus BJS Statistician ------------------------------------------------------------------ Highlights ------------ *Carjacking is defined as completed or attempted robbery of a motor vehicle by a stranger to the victim. It differs from other motor vehicle theft because the victim is present and the offender uses or threatens to use force. * Carjackings resulting in murder of the victim(s) are not covered by the National Crime Victimization Survey. According to FBI data, however, each year about 27 homicides by strangers involved automobile theft. These incidents may have been carjackings. * Between 1992 and 1996 an average of about 49,000 attempted or com pleted carjackings occurred in the United States each year. * About half of all carjackings were completed. * About 7 of 10 completed carjackings involved firearms, compared to 2 of 10 attempted carjackings. * Most carjackings, completed or attempted, did not result in injury to the victims. * All of the completed carjackings identified by the survey were reported to the police, while more than half of the attempts were reported. ---------------------------------------------------------------- An average of about 49,000 completed or attempted nonfatal carjackings took place each year in the United States between 1992 and 1996. In about half of the incidents, the offender was successful in taking the victim=s motor vehicle. In addition, although data on fatal carjackings are not available, about 27 homicides by strangers each year involved automobile theft. Some of these may have been carjackings. Carjacking, as defined in this report, is a completed or attempted theft in which a motor vehicle is taken by force or threat of force. It differs from other motor vehicle theft, which does not include incidents in which the offender used force or threats of force to obtain the vehicle. As explained in Meth- odology on page 5, only incidents in which offenders were strangers to the victim were included in this report, except where otherwise noted. Car- jackings that resulted in the murder of the victim(s) are not covered by the National Crime Victimization Survey (NCVS). Data about murders that involved automobile theft were obtained from the Uniform Crime Reports, collected by the Federal Bureau of Investigation. Characteristics of nonfatal carjackings Each year between 1992 and 1996, an average of about 2.5 per 10,000 persons age 12 or older were victims of a completed or attempted carjacking (table 1). ------------------------------------------------------------ Table 1. Characteristics of carjacking victims, by rates per 10,000 persons, 1992-96 Victim Annual average rates of attempted characteristic or completed carjackings per 10,000 persons Total 2.5 Age 12 to 24 2.5 25 to 49 3.6 50 or older 0.9 Sex Male 3.1 Female 1.9 Race White 2.0 Black 6.0 Other 2.2* Ethnicity Hispanic 6.1 Non-Hispanic 2.2 Marital status Never married 4.1 Married 1.4 Widowed 1.7* Divorced/separated 4.3 Household income Less than $15,000 2.9 $15,000 to $34,999 2.0 $35,000 to $49,999 3.2 $50,000 or more 2.4 Location of residence Urban 4.0 Suburban 2.4 Rural 0.7* Note: Based on NCVS data. Does not include carjackings involving murder. ---------------------------------------------------------- Ninety-two percent of carjacking incidents involved one victim, and 1% involved three or more victims. Generally the categories of people the most vulnerable to nonfatal carjacking were those observed to be the most vulnerable to violent crime overall. Men were more frequently victimized than women; blacks more often than whites; Hispanics, than non-Hispanics; and divorced, separated, or never-married persons, than those who were married or widowed. Urban residents were more likely to experience carjacking than suburban or rural residents. Persons over age 50 had lower rates than younger persons, but there were no clear patterns by household income. A weapon of some type was used in 83% of all carjackings (Highlights and figure 1). Firearms were used in 72% of completed carjackings and 22% of attempts. Overall, firearms were used in almost half of total nonfatal carjacking incidents. ------------------------------------------------- Figure 1. The offenders carried a gun in almost three-fourths of completed, nonfatal carjackings Carjackings, 1992-96 Weapon used Completed Attempted All 100% 100% No weapon 8.3 25.3 Firearm 71.9 22.1 Other weapon/unknown 19.8 52.4 ------------------------------------------------- Thirty-eight percent of carjacking victims offered no resistance to the offender (table 2). Nineteen percent used confrontational resistance, such as threatening the offender with a weapon, attacking the offender, chas- ing or trying to capture the offender, and other similar methods. A third of victims used nonconfrontational methods, such as running away, calling for help, trying to get attention, and other similar methods. Data were not sufficient to analyze the relative impact of various self-defense methods on carjacking completion or victim injury. ------------------------------------------------- Table 2. Measures that victims of carjacking took to protect themselves, 1992-96 Self-protective measure Percent of all carjackings Total 100% No resistance 38 Threatened/attacked with weapon 3* Unarmed confrontational resistance/a 16 Nonconfrontational actions/b 34 Other 9* *Based on fewer than 10 sample cases. a/Includes attacking the offender, chasing or trying to capture the offender, and other similar methods. b/Includes running away, calling for help, trying to get attention, screaming, and other similar methods. ------------------------------------------------- Most carjacking victims were not injured. About 16% of victims of attempted or completed carjackings were injured. Serious injuries, such as gunshot or knife wounds, broken bones, internal injuries, and loss of consciousness, occurred in about 4% of all carjackings. More minor injuries, such as bruises, chipped teeth, and other similar injuries, occurred in about 13% of all carjacking. Percent of victims of attempted or completed Type of injury carjackings Uninjured 84% Serious injuries/a 4* Minor injuries 13* *Based on fewer than 10 sample cases a/Includes gunshot or knife wounds, broken bones, internal injuries, and loss of consciousness. About 13% of surveyed carjacking victims were treated in hospital emergency departments or outside hospital settings, such as in doctors= offices or at the crime scene. About 3% of victims did not seek medical treatment even though they were injured. Percent of car- Medical jacking treatment victims Injured Emergency room, doctor's office, crime scene 13% Untreated 3 Uninjured 84% Carjacking incidents were about evenly divided between those committed by a lone offender and those committed y more than one offender (table 3). ----------------------------------------------------------- Table 3. Characteristics of carjacking offenders, 1992-96 Percent of completed Characteristic or attempted of carjackings and nonfatal carjackings Total 100% Number in incident 1 45 2 or more 55 Sex/a All male 97 Both male and female 3* Race/a White 19 Black 58 Other/b 16 More than one race 5* *Based on fewer than 10 sample cases. a/Combines characteristics of carjackers in single-offender and multiple-offender incidents. b/Includes Asians and American Indians. Respondents also may have incorrectly classified Hispanic offenders as belonging to "other races." ----------------------------------------------------------- Males committed 97% of carjacking incidents, and groups with both males and females committed 3%. Fifty-eight percent of carjacking incidents were committed by offenders whom the victim perceived to be black, and 19% were committed by offenders perceived to be white. In 5% of carjacking incidents, the victim(s) reported that multiple offenders of more than one race committed the crime. Of the daytime carjacking incidents, almost 2 in 3 were completed. Less than half of the nighttime incidents were completed. Time of occurrence All Completed Attempted Total 100% 50% 50% Day (6 a.m.- 6 p.m.) 100 64 36 Night (6 p.m.- 6 a.m.) 100 42 58 Most carjacking incidents occurred away from the victim=s home. Forty percent occurred in an open area, such as on the street (other than adjacent to the victim's home or the home of a friend or neighbor), near a bus, subway, or train station or near an airport (table 4). Twenty percent occurred in parking lots or near commercial places like stores, restaurants, gas stations, and office buildings. ------------------------------------------------------ Table 4. Places where carjackings occurred, 1992-96 Percent of carjacking incidents Characteristic of incident All Completed Attempted Total 100% 100% 100% Place of occurrence At or near victim's/friend's/neighbor's 26% 25% 26% Commercial place/parking lot/a 20 16* 25 Open area/public transportation/b 40 46 35 Other 13 13* 14 Distance from home At/near home 22% 25% 19%* 1 mile or less 22 27* 18* 5 miles or less 21 16* 26 50 miles or less 30 30 29 More than 50 miles 5* 3* 8* *Based on fewer than 10 sample cases. a/Includes stores, gas stations, office buildings, restaurants, and other commercial places. b/Includes on the street (other than in front of victim's, neighbor's, or friend's home), in parks, and on public transportation, such as in a bus or train station or in an airport. ------------------------------------------------------ Although carjackings often occurred away from the victim=s home, they usually did not occur very far away. About 65% of carjacking incidents occurred within 5 miles of the victim's home. About 5% of the carjackings happened more than 50 miles from the victim's home. All of the completed carjacking incidents identified by the survey were reported to the police. More than half of the attempted incidents were also reported, according to the victims. Percent of carjackings reported to the police All 79% Completed 100 Attempted 57 Carjacking victims received at least partial recovery of their property in 7 of 10 completed incidents. Fifteen percent of completed carjackings involved recovery of all property. Incidents with partial recovery most likely involved the recovery of vehicles either damaged or missing stolen articles. Percent of Recovery completed of property carjackings All 15% Part 56 None 27 Unknown 2* *Based on fewer than 10 sample cases. Thirty-five percent of completed carjacking incidents were reported to insurance companies. Most carjacking victims indicated that they either had no theft insurance or did not report the theft to their insurance company. Completed carjackings were less likely than completed motor vehicle thefts in general to be reported to insurance companies. Reported Percent of completed-- to Carjack- Motor vehicle insurance ings thefts Yes 35% 57% No or without insurance 63 41 Unknown 2* 2 *Based on fewer than 10 sample cases. ------------------------------------------------------ Box on carjackings by offenders known to victims ************************** Carjacking by persons the victims knew ************************** Although carjacking in this report is confined to robbery by strangers, NCVS data also describe motor vehicle theft by force by nonstrangers. Victim- Percent of violent offender motor vehicle relationship robberies, 1992-96 Stranger 72% Intimate or relative 15 Other 14 Strangers to the victims committed 72% of the carjacking incidents measured by NCVS. Intimates, such as spouses, ex-spouses, boyfriends, or girlfriends, or relatives committed 15%. The remainder of the victims either said they knew the offender as an acquaintance or did not report to the NCVS their relationship to the offender. ------------------------------------------------------ ------------------------------------------------------ Box on Federal cases ********************* Federal prosecution of carjacking ********************* Each year from fiscal year 1992 through fiscal year 1996, U.S. attorneys filed an average of 229 carjacking cases in U.S. district courts. In fiscal year 1996 U.S. attorneys declined to pursue Federal prosecution of 137 defendants charged with carjacking. These data are from the BJS Federal Justice Statistics Program. Current Federal law states that armed carjacking can be prosecuted as a Federal crime "as appropriate and consistent with prosecutorial discretion." (See P.L. 102-519 Sec. 101(b).) ------------------------------------------------------ ------------------------------------------------------ Box on infants in carjackings ********************** Infants taken during carjackings: A search of news reports ********************** One fear of carjacking is that an infant strapped into a carseat will be abducted. While the NCVS does not have data on carjackings in which infants were taken, a search of news accounts for 12 months in 1997-98 provided some information. Eight infants, ages 2 to 14 months, were kidnaped in carjackings, according to news reports. Most of these incidents occurred during the day. Half were at a service station, supermarket, or shopping center; half were on the street or in a parking lot. Carjackers attempted to force the driver from the car in two incidents and succeeded in one. Seven of the eight infants were found shortly after the incident, or the attempt was foiled. At the outset most offenders were unaware of the baby and within a short time left the infant and/or abandoned the car. The National Criminal Justice Reference Service conducted the Lexis-Nexis newspaper search for March 1997 through March 1998. The search covered more than 2,300 full-text information sources and more than 1,000 abstracted sources. Most carjackings with infant kidnaping had multiple references, which were consolidated. The search may have missed some incidents, particularly attempted carjackings and those not reported in newspapers or receiving only local coverage in small newspapers. ------------------------------------------------------ ------------------------------------------------------ Box on homicide and car theft ****************************** Homicides involving motor vehicle theft ****************************** The findings presented using NCVS data cover carjackings that did not result in murder. The Supplemental Homicide Reports from the FBI's Uniform Crime Reports include information on circumstances associated with murder. One of these circumstances is auto theft. It is not possible to determine how many of these murders with auto theft involved direct confrontation of the victim with carjacking as the motive. For example, the offender or offenders may have entered a home, murdered the occupants, and taken the car as an escape vehicle. Between 1992 and 1996 strangers committed 49% of the 272 murders involving auto theft. Of the nine murder/auto thefts with a victim under age 16, one involved multiple murders. Possibly a victim age 16 or older was driving a carjacked vehicle in which an occupant under 16 was murdered. There are insufficient data to conclude that these (or other) murder/auto thefts were carjackings. These data show, however, that carjacking involving murder is a rare event. If all murder/auto thefts committed by strangers between 1992 and 1996 were carjackings, there would have been an annual average of 27 homicides committed with a carjacking. ------------------------------------------------------ Methodology Except for the information concerning infants in carjackings and auto theft involving murder (see the boxes on page 4), this report presents data from the National Crime Victimization Survey (NCVS). The NCVS obtains information about crimes from an ongoing, nationally representative sample of households in the United States. NCVS data include both those incidents reported and those not reported to the police. Between 1992 and 1996 approximately 221,000 households and 430,000 persons age 12 or older were interviewed. Response rates between 1992 and 1996 ranged from 91% to 96% of eligible households and individuals. Carjacking is not one of the crimes measured every year by the NCVS because of an insufficient number of cases identified annually. To measure nonfatal carjacking over multiple years using NCVS data, cases were selected based on the following criteria: (a) a completed or attempted robbery occurred, (b) a car or other motor vehicle, such as a truck or any other motorized vehicle legally allowed on public roads or highways, was taken or an attempt was made to take such a vehicle, and (c) the offender or offenders were strangers to the victim(s). Because of the relatively small number of carjacking incidents occurring annually, findings from a sample may not fully measure the characteristics of carjackings. For example, although all of the completed carjacking incidents measured by the sample were reported to the police, we cannot conclude that completed carjackings are always reported to the police. Similar caution should be exercised in interpreting other findings. End of File 3/2/99 th revised 5/19/99 pm