Report on Cocaine and Federal Sentencing Policy

Chapter 5

Cocaine and Crime

A. INTRODUCTION

There is widespread belief that cocaine in general and crack cocaine in particular "causes crime to go up at a tremendously increased rate." See 134 Cong. Rec. S17,301 (Oct. 21, 1988) (statement of Sen. Helms, urging that crack cocaine offenses be subject to mandatory minimum penalties in part because "crack [cocaine] has been linked to violent crime"); 134 Cong. Rec. E2,701 (Aug. 10, 1988) (statement of Rep. Miller, expressing concern over the link between the crack cocaine trade and gang activity); 132 Cong. Rec. 31,330 (Oct. 15, 1986) (statement of Sen. Chiles); Hearing on Crack Cocaine, Permanent Subcommittee on Investigations of the Committee on Governmental Affairs, United States Senate, 99th Congress; C. Reinarman and H. Levine, "The Crack Attack: Politics and Media in America's Latest Drug Scare," in J. Best (Ed.), Images Typifying Contemporary Social Problems 117 (1989). See also, I. Wilkerson, "Crack's Legacy of Guns and Death Lives On," The New York Times A1 (Dec. 13, 1994). During debate about the Anti-Drug Abuse Act of 1986, for example, members of Congress expressed deep concern about increased crime related to crack cocaine. This chapter provides an overview of the current understanding of the connection between both powder and crack cocaine and crime. Sources reviewed here include empirical analyses, published and unpublished, and public testimony received by the Sentencing Commission. United States Sentencing Commission, Hearing on Crack Cocaine (Nov. 1993) (hereinafter "Commission Hearing").

Section B summarizes the limited conclusions drawn by researchers to date on crime and cocaine through a framework that has been widely recognized as helpful in understanding and analyzing the relationship between drugs and crime. Section C provides some context for assessing the association between cocaine and crime. This is done through analyses of the social context surrounding cocaine distribution, how violence associated with both powder and crack cocaine compares historically to violence associated with other "drug eras," and how crime associated with both powder and crack cocaine compares to that associated with other drugs.

There are at least two important limitations concerning the research relied on in this chapter and in research on the relationship between drugs and crime in general. First, conducting research in this area and drawing conclusions from it is complex and otherwise difficult. Determining, for example, whether trafficking in a specific drug has a causal relationship with crime requires studies that disentangle trafficking in that drug from all concurrently influencing factors. It also requires that conclusions based on a particular sample, at a particular time, and in a particular place, not be readily generalized to the broader population.

Second, in part because of the complexity, little reliable research is available on specific drugs and their relationships to criminal activity. Moreover, there is even less research available on the differences in varying forms of a single drug, such as crack and powder cocaine. This chapter thus relies on the handful of currently available studies that investigate cocaine and crime. The Commission recognizes, as should readers of this report, the limitations of the available research data.

While there is little doubt that an association between drugs and crime can be found, the literature on the drugs/crime connection still provides no consensus as to whether drug use causes crime, involvement in crime causes drug use, or other factors cause both. To inform policy better in this important area, the Sentencing Commission, in conjunction with Florida State University, has recently initiated an examination of causal relationships between drugs and violent crime. The study will distill the body of literature on drugs and violent crime and conduct independent research to build on currently available research. The study is expected to be completed in early 1996.

B. DRUGS, CRIME, AND THE TRIPARTITE FRAMEWORK

In 1985, Dr. Paul J. Goldstein of the University of Illinois School of Public Health described "a tripartite conceptual framework" for analyzing drug-related crime, especially violent crime. P. Goldstein, "The Drugs/Violence Nexus: A Tripartite Conceptual Framework," 14 Journal of Drug Issues 493 (Fall 1985). The Goldstein framework increasingly has been recognized by researchers and others as helpful in understanding the nature of drug/crime associations. See J. Inciardi, "The Crack-Violence Connection Within a Population of Hard-Core Adolescent Offenders," in M. de la Rosa, B. Gropper, and E. Lambert (Eds.), Drugs and Violence: Causes, Correlates and Consequences 92 (1990) (hereinafter "1990 Inciardi Adolescent Study"); K. Chin and J. Fagan, "Violence as Regulation and Social Control in the Distribution of Crack," in M. de la Rosa, B. Gropper, and E. Lambert (Eds.), Drugs and Violence: Causes, Correlates and Consequences 36 (1990) (hereinafter "1990 Fagan/Chin Study"); A. Reiss and J. Roth, Alcohol, Other Psychoactive Drugs, and Violence in Understanding and Preventing Violence (1993); J. Fagan, "Intoxication and Aggression" in M. Tonry and J.Q. Wilson Drugs and Crime (1990) (hereinafter "1990 Fagan Intoxication Study"). The Goldstein framework sets out three principal types of drug-related crime: systemic crime, psychopharmacologically driven crime, and economically compulsive crime. P. Goldstein, "Drugs and Violent Crime," Pathways to Criminal Violence 16, 24 (Neil A. Weiner et. al., eds., 1989) (hereinafter "1989 Goldstein Violent Crime Study"). Although this framework was developed with violent crime in mind, its economic-compulsive prong is useful and relevant in considering nonviolent drug-related crime as well.

1. Systemic Crime

Systemic crime arises out of the system of drug distribution. Id. at 30. It includes:

disputes over territory between rival drug dealers, assaults and homicides committed within dealing hierarchies as a means of enforcing normative codes, robberies of drug dealers and the usually violent retaliation by the dealers or their bosses, elimination of informers, disputes over drugs and/or drug paraphernalia, punishment for selling adulterated or phony drugs, punishment for failing to pay one's debts, and robbery violence related to the social ecology of copping areas. Id.

Systemic violence has been referred to as a means to achieve "economic regulation and control" in an illicit market. 1990 Fagan/Chin Study, supra note 4, at 36. As one expert at the Sentencing Commission hearing on crack cocaine explained regarding this type of crime, "[i]n an underground economy, you can't sue. So you use violence to enforce your breaches of contract or perceived breaches of contract." Commission Hearing, supra note 2, at 72 (testimony of Jerome H. Skolnick).

a. Empirical Findings on Crack and Powder Cocaine and Systemic Crime

As noted in Chapter 4, many retail powder cocaine distributors also distribute crack. Thus, pulling apart the systemic crime associated with crack cocaine versus powder cocaine is difficult, if not impossible. As one study noted, "it is the frequency of selling cocaine products, not just selling in its smokeable form, that seems to best explain violence in [cocaine] selling." See 1990 Fagan/Chin Study, supra note 4, at 27.

At the Sentencing Commission hearing on crack cocaine, a panel of noted researchers The panelists were Steven Belenko, Senior Research Fellow at the New York City Criminal Justice Agency; Jerome H. Skolnick, Professor of Law at the University of California, Berkeley; and Paul J. Goldstein, Associate Professor at the School of Public Health at the University of Illinois, Chicago Circle. addressed the specific topic of crack cocaine and its relationship to violent crime. The panel uniformly agreed that currently, "the primary association between [crack] cocaine and violence is systemic. It is the violence associated with the black market and distribution." Commission Hearing , supra note 2, at 67. Dr. Steven Belenko explained that such factors as the "volatile and jittery" nature of the early crack cocaine market, its tendency to attract younger, presumably more crime-prone sellers, and later attempts by organized dealer groups to exert control all led to an atmosphere in which participants in the crack cocaine trade were apt to "use . . . violence to maintain discipline, resolve disputes, and enforce control." Id. at 55-56.

The violent nature of the crack cocaine marketplace has been documented in three recent studies. A study of homicides in New York City during 1988 reported by Goldstein et al., P. Goldstein, H. Brownstein, P. Ryan, and P. Belluci, "Crack and Homicide in New York City, 1988: A Conceptually Based Event Analysis," 16 Contemporary Drug Problems 650 (Winter 1989) (hereinafter "1988 Goldstein et. al., Homicide Study"). found that of 118 crack-related homicides that were studied, 85 percent were systemic in nature. Id. at 664 (table 2). The study examined over 400 New York City homicides during 1988 and found that about 53 percent were "drug related"; of these, about 60 percent were related to crack. Id. at 662-663. Twenty-nine percent of the homicide perpetrators and 34 percent of victims were identified by authorities as drug traffickers, the "vast majority" of whom were considered to be "low level traffickers." Id. at 661. Another study found that street-level dealers, who typically carry smaller quantities of crack cocaine and money, were less likely to be involved with violence than dealers at a higher level in the distribution chain. See M. Klein, C. Maxson, and L. Cunningham, "'Crack,' Street Gangs, and Violence," 29(4) Criminology 623-650 (1991). The DEA has reported, however, that street-level crack cocaine sales may involve a heightened risk of violence because street sellers conduct business in uncontrolled situations and may be unfamiliar with their customers. See U.S. Department of Justice, Drug Enforcement Administration, DEA Drug Situation Report: Crack Cocaine iii, 10 (November 4, 1993) (draft). The study found that seven crack-related homicides were "multi-dimensional," with systemic being one of the dimensions. Id. at 675-78.

The 1990 Inciardi Delinquent Adolescent Study of "seriously delinquent" adolescent offenders in Miami from 1985 to 1988 1990 Inciardi Delinquent Adolescent Study, supra note 4, at 92. also found an association between crack selling and violent crime. The sample consisted of 611 adolescents who had committed at least ten FBI "index" offenses, The author reports that "index" offenses, in the FBI's Uniform Crime Reports, include criminal homicide, forcible rape, aggravated assault, robbery, burglary, larceny/theft, motor vehicle theft, and arson. Id. at 92. or 100 lesser crimes, in the preceding 12 months. A second criterion for the sample was that the subjects used some kind of illegal drug regularly at any time during the 90-day period prior to the study. Id. The study reported that 29.8 percent of the adolescents used crack cocaine regularly, and 29.3 percent used powder cocaine regularly. It also reported that those involved in dealing crack cocaine committed significantly more robberies than those who were not so involved. Id. at 104. However, the study reported that higher rates of crack use and distribution do not necessarily translate into higher homicide rates (except in Washington, DC). Id. at 107. The study suggested that "the current focus on crack-related violence may be more the result of a media event than an emergent trend." Id. at 105.

A 1990 study by Jeffrey Fagan and Ko-lin Chin found evidence that violence is associated specifically with the "economic regulation and control" of the cocaine marketplace. Id. at 36. The study compared results for crack and powder cocaine sellers and found that significant percentages of both regularly engaged in a range of violent interpersonal conflicts associated with selling (e.g., assaults to collect debts, fights with other sellers over drug quality). Id. at table 6. The study noted that any increased violence in the crack market was due to two factors:

First, crack selling was concentrated in neighborhoods where social controls had been weakened by intensified social and economic dislocations in the decade preceding the emergence of crack. Second, the rapid development of new drug-selling groups, following the introduction of crack brought with it competition. Accordingly, violence within new selling groups internally to maintain control and violence and externally to maintain selling territory . . . was more likely to characterize the unstable crack markets than more established drug markets and distribution systems. See 1990 Fagan/Chin Study, supra note 4, at 25.

Systemic violence also has been found in analyses of powder cocaine markets. For example, as Inciardi reports, in the late 1970s and early 1980s, Miami's cocaine distribution network experienced vast systemic crime. Prior to this period, Colombians had shipped powder cocaine to Miami, where middlemen distributed it locally or transhipped it elsewhere. See 1990 Inciardi Delinquent Adolescent Study, supra note 4, at 108. In the late 1970s, the Colombian drug kingpins moved to control the market without the middlemen and to take over cocaine distribution in South Florida. According to the study, this led to vastly increased systemic violence as territory was carved out among distributors. The murder rate rose to an all-time high of 621 murders (or 58.8 murders per 100,000 people) in 1981. Id. at 107. After the market stabilized and the Colombians gained control, the murder rate dropped by a third, down to a low of 33.2 murders per 100,000 people in 1987. Id. at 108 (table 9). As crack distribution increased, however, the murder rate rose again after 1987 to 42.5 murders per 100,000 in 1988 and 40.5 in 1989. Id.

2. Psychopharmacologically Driven Crime

Psychopharmacologically driven crime occurs when "individuals, as a result of short- or long-term ingestion of specific substances, become excitable, and/or irrational and exhibit violent behavior." 1989 Goldstein Violent Crime Study, supra note 5, at 24. In short, use of the drug directly affects behavior, one consequence of which is criminal conduct.

Goldstein cites as an example of psychopharmacologically driven crime his study of heroin-using prostitutes, who may behave more like robbers than prostitutes if they are experiencing withdrawal symptoms. In this state, the women reported "they might attack the client, take his money, purchase sufficient heroin to `get straight,' and then go back out on the street" to return to "regular" prostitution. Id. at 25.

Goldstein notes that drugs also may have a psychopharmacological effect if they are used to boost courage to commit crimes, either because they affect the brain in this manner directly or because the user expects the drugs to have this effect and, through a process of "self-fulfilling prophecy," they do. Id. at 26. In addition, psychopharmacologically driven violence may stem from drug use by the victim as well as the perpetrator. In other words, "drug use may contribute to a person behaving violently or it may alter a person's behavior in such a manner as to bring about that person's violent victimization." Id.

As discussed in Chapter 2, powder and crack cocaine contain the same active ingredients and thus the psychopharmacological effects of the two are qualitatively the same. The psychopharmacological effects of cocaine use, however, can differ dramatically as a result of the quantity used, the time period over which the use occurs, and the method of consumption (see Chapter 2).

a. Empirical Findings on Crack and Powder Cocaine and Psychopharmacologically Driven Crime

The limited evidence to date suggests that psychopharmacologically driven crime may be least important in explaining the association between crime and both crack and powder cocaine. With respect to violent crime, the 1990 Goldstein et al., Homicide Study found that only three of the 118 exclusively crack-related homicides in the study were psychopharmacological in nature, and in two of these three cases the victim precipitated the crime. The study concluded that there were another two psychopharmacologically driven homicides in which crack was involved. However, alcohol also was involved in these two cases, and overall, some 21 alcohol-only homicides were considered to be psychopharmacologically driven - considerably more than for any other drug - suggesting that alcohol may have played a significant role in these two crack-related cases. Id. at 664 (table 2), 665.

The 1990 Inciardi Delinquent Adolescent study found that only 5.4 percent of its sample of seriously delinquent adolescents - adolescents who commonly (but not necessarily exclusively or even primarily) used crack cocaine - reported "involvement" in psychopharmacologically driven violence at least once in the prior 12 months. 1990 Inciardi Delinquent Adolescent Study, supra note 4, at 98. Given that nearly 80 percent of the sample also reported involvement in "major felonies" during the same time period - a total of 18,477 such felonies committed by 611 adolescents in the 12-month time frame Id. at table 5. - the reported incidence of psychopharmacologically driven violence is relatively low.

A 1990 study by Fagan also generally concluded that "to date, there has been no systematic research linking crack cocaine use with increased [psychopharmacologically driven] violence." 1990 Fagan Intoxication Study, supra note 4, at 241, 257. Fagan went on to note, however, that "there is evidence of a sudden and precipitous depression following crack use." Id. He surmised this depression may be more causally related to subsequent economically compulsive crime than to psychopharmacologically driven crime. Id.

3. Economically Compulsive Crime

Economically compulsive crime is committed by persons who are financially driven to the criminal activity by financial needs brought about by drug consumption - for example, robbery that is committed by drug users "in order to support costly drug use." 1989 Goldstein Violent Crime Study, supra note 5, at 27. Goldstein notes:

Economically compulsive actors are not primarily motivated by impulses to act out violently. Rather, their primary motivation is to obtain money to purchase drugs. Violence generally results from . . . [s]uch factors [as] . . . the perpetrator's own nervousness, the victim's reaction, [the presence of] weaponry . . . and so on. Id.

a. Empirical Findings on Crack and Economically Compulsive Crime

A recent study by Inciardi and Pottieger J. Inciardi and A. Pottieger, "Crack-Cocaine Use and Street Crime," Journal of Drug Issues (forthcoming 1994) (on file with University of Delaware Center for Drug and Alcohol Studies) (hereinafter "Forthcoming Inciardi/Pottieger Users Study"). focused on the criminal activities of the users of crack cocaine. The study found that male "street users" users from neighborhoods with high rates of cocaine use engaged in a large number of criminal offenses, Id. at 15. the vast majority of which - more than 98 percent - were retail drug sales. Id. Most of these street users also reported that some of their living expenses and over 90 percent of their drug use were financed by crime, suggesting that street users rely on frequent, relatively small drug sales to support their crack cocaine habit. See id. at 29.

This is not to say, the authors noted, that street users did not engage in other criminal activity to generate cash. The study found, in fact, that 48 percent of the men and 62 percent of the women committed, on average, one "petty property crime" (e.g., shoplifting) per week, and some 69 percent of women users "were trading sex for money or drugs, or helping a prostitute partner do so." Id. at 18-19. The authors also reported that "a significant minority of the men were engaged in fairly high numbers of violent or potentially violent offenses, most commonly as an adjunct to their drug business offenses." Id. at 19. Relatively speaking, however, the criminal conduct of the street users was tilted heavily toward retail crack cocaine selling.

The authors' profile of these offenders as primarily users who sold crack to support their crack consumption - as opposed to sellers who used crack incidentally to their trade - appeared to be supported by a finding that while every male subject (and 94% of female subjects) reported making some retail drug sales, no subjects reported manufacturing or wholesaling crack cocaine. See id. at tables 4 and 6. The study did find, though, that male users in the street user sample who were "engaged in fairly high numbers of violent or potentially violent offenses . . . most commonly [committed such crimes] as an adjunct to their drug business offenses," suggesting a largely systemic component. Forthcoming Inciardi/Pottieger User Study, supra note 44, at 19.

The fact that many retail crack cocaine sellers are users who deal primarily to finance their consumption of crack is supported by other studies as well. About 61 percent of crack cocaine dealers in one Detroit study cited the desire to consume crack as the principal motivation for their dealing. T. Mieczkowski, "Crack Distribution in Detroit," 17 Contemporary Drug Problems 9, 23 (1990). In a Miami study, 80 percent of delinquent youths who used crack cocaine also sold it. J. Inciardi and A. Pottieger, "Kids, Crack, and Crime," 21 (2) The Journal of Drug Issues 257, 260 (1991).

A different analysis of crack users in drug treatment - "treatment sample" - suggested that these crack users are relatively less likely to have engaged in retail drug sales and more likely to have committed "large numbers of petty property crimes" prior to treatment. Id. at 19. The authors surmised that the difference in retail drug selling activity by the street and treatment samples could be due to the fact that:

the street sample consisted of the crack users who happened to be in good locations in which to support their crack use and other expenses by dealing. The treatment sample, on the other hand, may be more representative of the customer base of these dealers, and hence more representative of all crack users. Id. at 24.

The 1988 Goldstein et al. Homicide Study, discussed above, concluded that eight of the 118 exclusively crack-related homicides in the study were economically compulsive crimes. Six of the eight murders involved the murder of elderly persons during a robbery or burglary. One involved an attempted robbery of one crack user by another. The last murder allegedly was victim-precipitated; the victim allegedly was murdered trying to steal auto parts to support his crack habit. 1988 Goldstein et. al. Homicide Study, supra note 14, at 666-67.

As discussed earlier, there is evidence of increased involvement in prostitution by crack users. Women often trade sex for money or drugs, and some men become "pimps" to support their crack habit. K. Chin and J. Fagan, The Impact of Crack on Criminal Careers: Crime and Drug Involvement Following Initiation Into Cocaine Smoking (Aug. 1992) (unpublished, on file with Rutgers University School of Criminal Justice) (hereinafter" 1992 Chin/Fagan Study"). However, studies further indicate that prostitution is an economically compulsive crime for women who use both crack and powder cocaine (see Chapter 3).

4. Crime Indirectly Related to Crack

The Goldstein tripartite framework seeks to explain crime that is drug related, either because the crime is an adjunct to the unregulated marketplace (systemic), is a means to support drug consumption (economically compulsive), or occurs because of the drug's direct (or assumed) psychopharmacological effects on behavior. The tripartite framework, however, does not answer the question as to whether drug sellers, including cocaine sellers, have a tendency to use violence outside the drug context. Nor do other data at this point appear to offer a clear explanation of this association.

Researchers have speculated, however, that nondrug violence may be "intensified" See also 1990 Fagan/Chin Study, supra note 4, at 36. ("The crack market apparently has intensified the social processes that sustain both drug-related and other violence.") by the cocaine marketplace (and specifically the crack cocaine marketplace) because systemic violence creates a setting in which violent behavior generally is deemed acceptable. S. Belenko, J. Fagan, and K. Chin, Typologies of Criminal Careers Among Crack Arrestees (Nov. 1989) (hereinafter "1989 Belenko et. al., Study"); Commission Hearing, supra note 2, at 59 (testimony of Steven Belenko). Others point to the socioeconomic status of innercity neighborhoods as contributing to the extension of market violence to nondrug settings (see Section C, infra). Nonetheless, empirical studies conducted to date tend to find an association between crack cocaine involvement and the commission of other kinds of crime. This is true regardless of whether involvement is gauged by using or selling crack cocaine.

In one such study, Steven Belenko et al.1989 Belenko, et al., Study, supra note 59. examined a group of New York City "crack arrestees," an undifferentiated group of crack cocaine users and sellers. Overall, the study found "both an increased incidence of violent arrest post-[crack] initiation for new offenders and an accelerated rate of violent arrests for those with prior records of violence." Id. at 21. The study concluded that the arrestees' increased violence was "not limited to the context of the drug transaction," but rather could occur in other settings. Id. at 25.

The Chin and Fagan 1992 Chin/Fagan Study, supra note 57, at 13-14. study, discussed above, was consistent with Belenko but contained a noteworthy refinement. The study distinguished between samples of crack cocaine "users" and "users/sellers" drawn from two New York City neighborhoods with high concentrations of crack cocaine activity. (The "users/sellers" category was denominated as such because the authors were unable to identify sellers who had not also used crack cocaine.) Id. at 5.

The authors found that following involvement with crack cocaine, users reported significant increases in aggravated assault, theft, and, among women, prostitution. Id. at 11. The authors also reported, however, that "no users reported initiation into any form of crime following crack initiation. Instead, it appears from this study that crack intensifies the behaviors in which users already were involved." Id. at 16.

The picture among users/sellers was somewhat different. Female users/sellers, who typically held only low-level trafficking positions, also reported increased prostitution following crack involvement; but male users/sellers, in contrast to users, reported significant increases in crime only with respect to selling stolen goods, Id. at 12. and their commission of burglaries appeared to drop. See id. at table 2.

On the other hand, while the data generally did not show that users/sellers increased their commission of violent and other crimes following crack initiation, "[users/sellers] were [already] extensively involved in crimes both within and outside the context of drug selling prior to initiation into crack." Id. This finding led the authors to conclude that "processes of social or self-selection seemed to attract active offenders into [that] marketplace." Id. In short, in the authors' view, the direct effect of crack on violent behavior seems to be less clear because of the users'/sellers' prior involvement in these behaviors and their general participation in the often violent world of drug selling. Id. at 16-17.

In their study, Chin and Fagan found that crack and powder cocaine both attracted younger people to drug selling and violence. They found that "arrest and conviction data suggest that violence and participation in drug selling are more strongly associated with crack than with cocaine [powder]."

C. COCAINE IN CONTEXT

This section provides additional context for evaluating the crime associated with cocaine.

1. The Social Context of Cocaine Distribution

All three panelists testifying on the association between crack cocaine and violence at the Sentencing Commission hearing stressed that crack/crime associations cannot be assessed in isolation from the social environment in which the marketplaces for these drugs occur. Commission Hearing, supra note 2, at 80. Dr. Skolnick stressed the importance of the varying gang cultures in which cocaine trafficking, including crack cocaine, is often a part. He observed that it is "the underlying culture of the gangs in a particular area that accounts for the violence more than anything else." Id. at 70. Dr. Skolnick's observation appears supported by a recently released study conducted for the National Institute of Justice. E. Walsh, "Chicago Street Gang Study Shows Fearful Toll of Powerful Weapons," Washington Post A4 (Nov. 29, 1993) (citing to study conducted by Carolyn Rebecca Black and Richard Black). That study concluded that gang turf battles in many areas were more likely to lead to homicides than drug trafficking disputes.

Dr. Belenko pointed to a range of concurrent non-cocaine forces that he indicated undermine a conclusion that cocaine in general and crack cocaine in particular cause crime:

[W]hile the crack subculture can be characterized as more violent and crime-involved compared with previous or parallel drug subcultures, the reasons for this are quite complex and probably not a function of any psychopharmacological effects. Thus, the media and public fears of a direct causal relationship between crack and other crimes do not seem to be confirmed by empirical data. Rather, the levels of violence and crime associated with crack appear to reflect parallel and other interactive forces that are related to the relative immaturity and volatility of the crack markets, the ages and types of persons initially attracted to crack distribution, the increasing social and economic disorganization of the nation's inner cities beginning in the 1980's, and the mounting proliferation of more powerful guns, as well as a spread of cheaper powder cocaine during the same period of time. Id. at 59.

Other researchers have made similar observations about the importance of non-crack factors. Socioeconomic factors, for example, are thought by many to impact directly on the drug/violence relationship. Some sociologists theorize that deviant behavior is more likely to occur in a situation in which individuals lack access to legitimate means to achieve their economic goals. See 1990 Fagan Intoxication Study, supra note 4, at 274. Others postulate that "in conditions in which law and governmental social control are least developed, violence would be more evident as a form of social control." See 1990 Fagan/Chin Study, supra note 4, at 13. The 1990 Fagan and Chin study discussed these theories in relation to the crack economy in the innercity.

Fagan and Chin considered crack cocaine development during a concurrent decline in the lawful economy of innercity neighborhoods. Citing evidence of heavy innercity job loss during a time of job creation in surrounding suburbs and the fact that small-scale sellers were able to participate in the income-generating crack cocaine market, the authors observed that crack cocaine distribution attracted participants at a time when economic and social counterweights to the underground economy were seriously diminishing. 1990 Fagan/Chin Study, supra note 4, at 10-12.

Noting "that the vast majority of [residents] in inner-city communities are not cocaine or heroin abusers or criminals," Bruce Johnson et al. similarly found that such factors as the prospects of employment in the crack trade for young persons "who most likely would be otherwise unemployed" played a role in expanding "the criminal underclass subculture." B. Johnson, T. Williams, K. Dei, and H. Sanabaria, "Drug Abuse in the Inner City: Impact of Hard-Drug Use and Sales on Low Income Communities," in M. Tonry and J. Wilson (Eds.), 13 Crime and Justice, An Annual Review of Research 9-68 (1990) (hereinafter "1990 Johnson et al. Inner City Study"). In sum, whatever the precise effects of social and environmental factors, a number of researchers stress their relevance in considering both the rapid development of crack cocaine and crack's association with crime.

2. Cocaine and Other Illicit Drug Markets

The association between drugs and crime is not unique to cocaine. Research previously has found associations between violent crime and marijuana, heroin, and other drug trafficking. See authorities cited in 1992 Chin/Fagan Study, supra note 57, at 4. Research conducted since the 1920s has suggested "that while the use of . . . [illicit] drugs does not necessarily initiate criminal careers, it tends to intensify and perpetuate them." Forthcoming Inciardi/Pottieger Users Study, supra note 44, at 5.

Few researchers who have explored cocaine/crime associations have also directly compared the associations of crime to other drugs. Researchers who have made such comparisons paint a somewhat mixed picture. As stated above, the Goldstein et al. Homicide Study found that 60 percent of drug-related homicides in New York City in 1988 were related to crack cocaine. However, because crack cocaine was a particularly popular drug during this period, this finding by itself sheds limited light on crack's relative association with drug-related violence.

The 1990 Inciardi Delinquent Adolescent Study and a companion study J. Inciardi and A. E. Pottieger, supra note 53, at 257. suggest a more definite answer. These studies compared crime patterns of "seriously delinquent" adolescent offenders depending on the offenders' "proximity to the crack market." The studies concluded that proximity to crack trafficking correlated with increased commission of major felonies and property crimes.

In particular, it should be noted that these data suggest that it is not drug sales in general but specifically the crack business which is so highly problematic. . . 86 percent of the non-crack business group were selling some drug, averaging around 200 sales per year. But the involvement of this group in major felonies and petty property crime was distinctly lower than that of youths with even minor involvement in the crack business, let alone compared to that of crack dealers. Id. at 268.

This characterization is consistent with testimony of Steven Belenko at the Sentencing Commission's crack hearing. Commission Hearing, supra note 2, at 49-60. Dr. Belenko stated that he had analyzed arrest data for crack cocaine sellers and determined that, relative to powder cocaine sellers, crack cocaine sellers had higher arrest rates for both "nondrug and violent crimes." Id. at 57.

The Commission's own data on federal cocaine offenders suggest that crack cocaine distributors are more violent than most other federal drug offenders. Federal crack cocaine offenders are more likely to possess a weapon and also more likely to have an extensive criminal record. (See Chapter 7 for a more detailed discussion of the characteristics of federal drug offenses and offenders.)

Cutting the other direction, perhaps, are findings in the 1990 Fagan/Chin Study. 1990 Fagan/Chin Study, supra note 4. This study analyzed systemic violence engaged in by drug sellers from two New York City neighborhoods with high concentrations of crack cocaine selling. Id. at 13-14. Noting that sellers in the study frequently sold more than one drug, the study found that retail crack cocaine sellers reported no more systemic violence than marijuana or heroin sellers. Id. at 25. The study found that those who sold powder cocaine in these neighborhoods - whether with crack cocaine or other drugs - reported the highest levels of systemic crime. Id. at 27.

The Inciardi/Pottieger User Study compared economically compulsive crime committed by crack users in Miami with that committed by a comparable sample of heroin users a decade earlier. As noted, this study found that more than 98 percent of the crimes committed by male "street" users of crack cocaine consisted of small retail drug sales; less than two percent were property or other crimes. In contrast, the authors found that "dealing represented 51 percent of total offenses for male heroin users, among whom another 34 percent of all crimes were thefts and other property crimes." Forthcoming Inciardi/Pottieger Users Study, supra note 44, at 17. These data show a distinction between the economically compulsive crime most associated with the study's sample of crack cocaine users (retail drug sales) and that associated with the sample of heroin users (a broader mix of drug and property crimes).

3. Violence Associated with the Current Cocaine "Epidemic"

At the Sentencing Commission hearing, Dr. Goldstein commented that systemic violence is not unexpected in a newly developing drug market such as crack cocaine:

Systemic violence fluctuates with phases of the illicit market economy. Rates of homicidal violence were high when a new market was being forged for powder cocaine. Wars between Colombian and Cuban syndicates for control of middle-level cocaine distribution contributed substantially to rising homicide rates in the late 1970s and early 1980s. When these wars were over, even though there was plenty of cocaine on the streets in the mid-1980s, homicide rates declined. The peak level of homicidal violence caused by the crack wars is similar to the peak caused by the powder cocaine wars which is, in turn, similar to the peak caused by the alcohol wars during prohibition. Id. at 67.

Whatever conclusions are drawn about current levels of systemic violence in the crack cocaine market relative to levels for the current powder cocaine market, researchers have tended to agree that, from a historical perspective, crack cocaine is not unique. Dr. Goldstein testified that the national homicide rate (based on the number of homicides per 100,000 population) had "changed very little over the last 25 years." In 1992, he stated, the homicide rate was lower than in 1980, when systemic violence arising out of the newly developing powder cocaine market was about at its peak, and lower than in 1933, at the end of prohibition. Id. at 65.

D. THE DRUGS/VIOLENCE TASK FORCE

In June 1993, in Washington, D.C., the Sentencing Commission held a Symposium on Drugs and Violence in America. One conclusion of the symposium was, as stated earlier, that the currently available research data on the relationship between drugs and violence is limited. As a result, the Commission, together with the School of Criminology and Criminal Justice at Florida State University, is now sponsoring a task force to acquire a better understanding of the drugs/violence relationship.

The task force plans to conduct an in-depth examination of the issues related to the drugs/violence relationship by bringing together the accumulated knowledge and expertise of state and municipal leaders, academia, related federal agencies, Congress, criminal justice professionals, and concerned citizens. This expertise will be used to examine existing research and other information and to oversee several research projects aimed at clarifying specific matters of concern. The task force will present findings and policy recommendations that will help guide the response to drugs and violence in the future. The task force is expected to issue its report in early 1996.

United States Sentencing Commission