II. Findings
The following tables and figures report the arrest and conviction trajectories of members of the marijuana 2003-2004 cohort from their arrest in 2003 or 2004 until June 2011. The unit of analysis is the individual, unless otherwise specified.
Arrests
Arrests are not a valid measure of criminal conduct. In the aggregate, arrest data may say as much or more about law enforcement practices, including where and when the police work, as they do about the actions of people who are arrested.[56] Nevertheless, because arrest records are so commonly cited in criminal analyses, we have included a brief overview of the arrests of cohort members during the study period. Table Two displays the number and type (misdemeanor or felony) of arrests occurring after the marijuana arrest terminating in the initiating MJACD.
As shown in Table Two, nearly 4 in 10 (38.0 percent) of the MJACD cohort had no subsequent arrests as of mid-2011. Another 29.1 percent (8,493 individuals) had only misdemeanor and no felony arrests, including 4,161 who had only 1 subsequent misdemeanor arrest. Of the total 9,571 individuals who had 1 or more felony arrests (32.8 percent of the cohort), approximately half (4,742 individuals) had a single felony arrest.
Subsequent Convictions
Figure 2 presents a graphic overview of the number and nature of the subsequent misdemeanor and felony convictions of members of the cohort from the time of their marijuana possession arrest in 2003 or 2004 through June 2011. In the tables and figures that follow, we drill down into these numbers to better understand their significance.
Click image to enlarge.
As shown graphically in Figure 3, four out of five (80.3 percent) of the members of the marijuana 2003-2004 cohort, or 23,418 out of the total 29,147 individuals, had no misdemeanor or felony convictions over the 6.5 to 8.5-year period they were tracked, and 9 out of 10 (90.3 percent) had no felony convictions.[57]
Table 3 shows that most cohort members with subsequent convictions did not have extensive criminal records. [58] For example, 1,695 individuals had a single misdemeanor conviction and no felony convictions. Another 1,202 were convicted of 2 or more misdemeanors and no felony convictions. That is, 2,897 individuals—9.9 percent of the entire cohort—had subsequent criminal records limited to misdemeanor convictions. Almost three-quarters (73.2 percent) of those who had felony convictions—2,072 out of 2,832 individuals— had a single felony conviction (not including any misdemeanor convictions). However, 760 individuals—2.6 percent of the cohort—accumulated more extensive records consisting of 2 or more felonies, including 38 who had 4 or more felony convictions. Half (51.2 percent) of all cohort members with felony convictions had no misdemeanor convictions.
In addition to tracking the number of subsequent criminal convictions, we also sought to determine the nature of the felony crimes for which cohort members were convicted. A substantial portion of the individuals with later felony convictions were convicted exclusively for drug felonies. Among the 2,832 cohort members convicted of one or more felonies, 1,114 (39.3 percent) were only convicted of drug felonies. Among those whose felony convictions were limited to drug offenses, 77.5 percent (864 individuals) had only a single drug felony conviction. Among those convicted of drug felonies, all but two were convicted of drug crimes below class A felony offenses.[59]
Convictions for violent felonies were relatively uncommon. [60] As shown in Table Five, only 3.5 percent of the entire cohort (1,022 individuals) had one or more violent felony convictions over the 6.5 to 8.5-year period they were tracked. Among them, 912 individuals (3.1 percent of the total cohort) had a single violent felony conviction. (Looked at differently, among all cohort members who had violent felony convictions, 82.9 percent had only one such conviction.) Only 110 individuals—0.4 percent of the cohort—had more than one violent felony conviction. Some individuals convicted of violent felonies were also convicted of nonviolent felonies. For example, of the 912 persons with a single violent felony conviction, 225 had nonviolent felony convictions as well. [61]
We constructed the category of “injurious misdemeanors” to capture convictions for lower level crimes that involved or risked physically harmful or offensive conduct. The category includes offenses such as misdemeanor assault (e.g. intentionally causing physical harm to another that does not reach the felony threshold of “serious physical harm,” e.g. a slap in a bar), weapons possession (e.g. possession of a gravity knife), or misdemeanor sexual offenses (e.g. unwanted touching of a stranger’s buttocks on the subway) (see Methodological Appendix for full list of offenses included in this category).
As shown in Figure 4, in addition to the 80.3 percent of the group (23,418 individuals) who had no subsequent criminal convictions at all, 6.1 percent (1,770 individuals) had criminal convictions that did not include injurious misdemeanors or violent felonies. Another 10.1 percent (2,937 individuals) had one or more injurious misdemeanors, but no violent felony convictions. [62]
Among the members of the cohort who were subsequently convicted of an injurious misdemeanor, the most common convictions were, in order of frequency:
- Criminal trespass in the second degree
- Driving while intoxicated
- Assault in the third degree
- Criminal possession of a weapon in the fourth degree
- Criminal contempt in the second degree or bail jumping
- Unauthorized use of a vehicle in the third degree [63]
Our aim in this study was to ascertain whether a significant percentage of people who enter the criminal justice system for misdemeanor marijuana possession go on to be convicted of serious felony offenses. We found that within 6.5 to 8.5 years after the initial marijuana possession arrest, 3.5 percent had one or more violent felony convictions, and another 6.5 percent had felony convictions for nonviolent offenses.
Assessing the significance of 3.5 percent as the proportion of the cohort who have subsequent violent felony convictions is necessarily a value judgment. Some people might think it an extremely large number, others might think it extremely small. Interpretation of any statistic always involves substantive and subjective policy and value judgments.
There is no readily available data on the rates of felony or violent felony conviction for comparable demographic groups in New York City who do not enter the criminal justice system on marijuana charges.[64] But the rate of felony and violent felony conviction among this group of first-time marijuana arrestees appears to be lower than the rate of felony conviction for the national population, taking into account age, gender, and race.[65]
One major national study, for example, estimated that the cumulative risk of imprisonment by the age of 33 (that is, being sentenced to a prison term for any felony conviction) for the cohort of men born between 1965-1969 was 5.3 percent for white men with no college education and 30 percent for black men with the same age and education profile.[66] Using a cross-sectional approach, another study estimated that over 6 percent of the adult non-black population had a felony conviction in 2010, whereas 25 percent of the black adult male population had a felony conviction in 2010.[67]
[56] K. Babe Howell, "Broken Lives from Broken Windows: The Hidden Costs of Aggressive Order-Maintenance Policing," NYU Review of Law & Social Change.
[57] However, there were also four individuals who had zero felony convictions, but a single “Youthful Offender Adjudication” where the top charge was a felony crime (see Youthful Offender Procedure, New York Criminal Procedure Law Article 720).
[58]See Methodological Appendix. Among those with no criminal convictions, there were 7,719 individuals who had convictions of non-criminal violations or infractions.
[59] Of the two cohort members who were convicted of Class A drug felonies, one had a single felony conviction for a class A drug offense and no other felony convictions and the other had two felony convictions, one of which was a class A drug felony; neither individual had violent felony convictions.
[60] Violent felony offenses include the major street crimes such as most charges of robbery, burglary, murder, rape, serious assault, and also many sex offenses and weapons possession charges. See Methodological Appendix for definition of violent felony offense.
[61] Individuals convicted of felonies may also have been convicted of misdemeanors.
[62]Individuals convicted of non-injurious misdemeanors but no violent felonies may also have had other non-injurious or nonviolent criminal convictions.
[63] Using the unit of analysis of conviction, and not the individual, over 90 percent of all the convictions for injurious misdemeanors among the cohort were for these offenses.
[64] Nor do we have detailed data on these arrestee’s educational background or socioeconomic status; our descriptive data is limited to the age, sex, and race or ethnic indicators at the time of arrest.
[65] See Table One for the demographic profile of the MJACD cohort.
[66] Bruce Western, Punishment and Inequality in America (New York: Russell Sage Foundation, 2006), pp. 26-27. Of course these numbers underestimate the cumulative risk of a felony conviction for this age cohort since not all felony convictions result in a prison sentence.
[67] Sarah Shannon et al., Population Association of America, "Growth in the U.S. Ex-Felon and Ex-Prisoner Population, 1948 to 2010," April 2011, http://paa2011.princeton.edu/papers/111687 (accessed October 16, 2012); Christopher Uggen, Jeff Manza and Melissa Thompson, "Citizenship, Democracy, and the Civic Reintegration of Criminal Offenders," Annals of the American Academy of Political and Social Science, vol. 65, no. 1(2006), pp. 281-310. The cross-sectional estimates are approximating the number of people in the entire adult population of each demographic group that has a felony conviction in a given year.