Opening Pandora’s Box: How Does Defendant Race Influence Plea Bargaining?

Abstract

Minority overrepresentation in the criminal justice system has long been an important topic of research and policy debate. In New York City, recent changes in the Rockefeller Drug Laws and the controversy around police stop-and-frisk practices have placed an even greater emphasis on the need for studying the possible impact of defendants’ race and ethnicity on criminal justice outcomes. Relatively little contemporary research, though, examines plea-bargaining outcomes. Using unique data on misdemeanor marijuana cases, this study examines the impact of defendants’ race on prosecutors’ decisions to make (a) plea offers for a lesser charge and (b) sentence offers for non-custodial punishments. Preliminary findings indicated that black defendants were less likely to receive reduced charge offers, and both black and Latino defendants were more likely to receive custodial sentence offers. However, these disparities were largely explained by legal factors, evidence, arrest circumstances, and court actor characteristics, though black defendants were still more likely to receive custodial sentence offers after including these controls. No differences were found between white and Asian defendants. Implications for research and prosecutorial practices are discussed.

Keywords:

• race and ethnicity
• prosecution
• plea bargaining
• evidence

Funding

This study was supported by grant 2011-DJ-BX-0038 from the National Institute of Justice, Office of Justice Programs, US Department of Justice.

Acknowledgements

Requests for additional information about this study should be directed to [email protected]. Points of view expressed in this report are those of the authors and do not necessarily represent the official position of the US Department of Justice or the views of the New York County District Attorney’s Office.

Notes

¹ The Uniform Crime Reports’ definition of “black” includes “black Hispanics.”

² Structured interviews conducted with 16 DANY prosecutors as part of a broader research project also suggested variation in the awareness of and adherence to charging guidelines, so there may also be important prosecutor-level variation in racial disparities.

³ The Rockefeller Drug Laws are the statutes dealing with the sale and possession of narcotics in the New York State Penal Law, named under Governor Nelson Rockefeller who signed them in 1973. The statutes carried a minimum of 15 years to life in prison, and a maximum of 25 years to life in prison for selling two ounces (57 g) or more of heroin, morphine, opium, cocaine, or cannabis, or possessing four ounces (113 g) or more of the same substances. In April 2009, these statues were revised to remove the mandatory minimum sentences and to allow judges to sentence individuals convicted of drug offenses to treatment or to shorter sentences.

⁴ The sample excluded 8,343 cases disposed of as “adjournments in contemplation of dismissal” (commonly known as ACDs) because these cases are sealed upon dismissal and therefore not available for research. ACD is an agreement between the district attorney’s office and the defense to have a case adjourned with a view to dismissal in a six-month or one-year period, if there has been no arrest for a new offense (see New York Criminal Procedure—Article 170—§170.55 Adjournment in Contemplation of Dismissal, and §170.56 Adjournment in Contemplation of Dismissal in Cases Involving Marijuana). A total of 5,798 case dismissals and seven acquittals were also excluded. The sample was also limited to defendants 16 years of age or older.

⁵ The distinction between class A and B misdemeanors is also important because § 221.40—Criminal sale of marihuana in the fourth degree is a class A misdemeanor and § 221.10—Criminal possession of marihuana in the fifth degree is a class B misdemeanor in New York State. The sample contained 84.2% of cases involving the § 221.10 charge.

⁶ Using drugs and sale of drugs were the most common arrest behaviors. Other activities occurred less frequently and include: whether drugs were spotted (not a sale); whether a warrant was executed for a search of drugs; whether a predicate stop was made (i.e. the police officer stopped the defendant for reasons other than drugs); or whether “furtive movements” resulted in arrest (i.e. behaviors deemed suspicious by police officers, which usually includes descriptions of the defendant running away or fidgeting).

⁷ Targeted police office activity includes: (1) undercover buy and bust (undercover officer buys or attempts to buy narcotics from defendant or those with whom defendant is arrested); (2) observation point (one police officer is observing area from a fixed location and when he sees suspicious activity radios to his field team; the team conducts actual stops, investigations, and arrests; (3) vertical sweep (vertical patrol of building, either NYCHA (public) housing or private homes participating in the Trespass Affidavit Program); and (4) prior investigation.

⁸ Assigned panel attorneys (pursuant to Article 18(b) of the County Law) have provided legal services to indigent defendants within the Bronx and New York County Criminal courts since 1966. They are private attorneys who are compensated for representing indigent clients and they are assigned matters when a conflict prohibits institutional providers, such as The Legal Aid Society, from providing representation.

⁹ Manhattan has the greatest income inequality by race of all five of New York City boroughs, with median household income for whites being roughly three times higher than for Latino and black households in the city.

¹⁰ Additional models (not reported) were also estimated examining the influence of different individual measures of prior criminal behavior and of socioeconomic proxies. Models that included only prior arrests produced similar estimates for race and ethnicity, whereas models that excluded prior arrest and only include prior prison and violent felony conviction produced slightly larger estimates of racial disparity. In models that omitted socioeconomic proxies like neighborhood SES, racial differences increased notably for both dependent variables.

¹¹ Preliminary analyses of missing data patterns were performed using the Missing Values Analysis add-on for SPSS. Multiple imputation was subsequently implemented using the mi impute command within STATA12 (commercially available from StataCorp). We created m = 40 multiply imputed data-sets and analyzed them using logistic regression within STATA’s mi estimate environment, which pools estimates from these multiple data-sets using specific rules proposed by Rubin (1987).

¹² We also explored two-level hierarchical linear modeling (HLM) with prosecutor variables added at level two of the analysis. Unconditional models examining variance components suggested that a statistically significant amount of variation in charge offers existed across prosecutors (²= 6.94; df = 1) but the same was not true for sentence offers (²= 1.21; df = 1), and once individual predictors were included the between-prosecutor variance was reduced to non-significance for both dependent variables. Because HLM does not accommodate our multiple imputation approach and because the between-prosecutor variance was non-significant when our individual predictors were included, we limit our discussion of findings to models that include standard errors corrected for clustering among prosecutors. Substantive findings were very similar for both approaches and results of the additional HLM models are available from the authors by request.

¹³ A two-level HLM analysis of charge offers produced equivalent findings.

¹⁴ Additionally, to explore the extent to which drug sale (class A misdemeanors) cases were influencing the results, we estimated models predicting charge offer, while excluding 190 (15%) class A misdemeanor cases. Results changed only nominally, whereby blacks were 1.56 times more likely than whites to receive a more punitive charge offer, but the relationship is still marginally significant (p = .06). Outcomes for Latinos (odds ratio = 1.16) and Asians (odds ratio = .76) remained statistically non-significant.

¹⁵ As a robustness check, the full model predicting custodial sentence offer was re-estimated excluding drug sale class A misdemeanor cases that may be driving plea offer outcomes. Results showed greater racial and ethnic disparities, with blacks being 2.35 times more likely than whites to receive a custodial sentence offer. The difference between Latinos and whites became statistically significant, whereby Latinos were 1.69 times more likely to receive a custodial sentence offer (p = .02). The difference between Asian and white defendants remained non-significant.

¹⁶ All supplemental analyses are available from the author by request.

¹⁷ According to the DANY plea guidelines, “plea bargaining is most appropriate for a defendant’s first arrest. On subsequent arrests of the defendant, the defendant ordinarily should be required to plead guilty to the top provable charge” (Plea Guidelines: Introductory Principles, Principle 3).

Additional information

Notes on contributors

Besiki Luka Kutateladze

Besiki Luka Kutateladze served as a research director in the Prosecution and Racial Justice Program at the Vera Institute of Justice at the time of the project and was the principal investigator on this study. Previously, he has worked on various rule-of-law and justice projects supported by the United Nations, the UK Department for International Development, and the World Justice Project. At the time of publication, he became an associate director for Research at the Institute for State and Local Governance of the City University of New York. His areas of interest include race, prosecution, punitiveness, rule of law measurement, and governance performance indicators.

Nancy R. Andiloro

Nancy R. Andiloro served as a research associate at the Vera Institute of Justice. She spent two years assigned to the New York County District Attorney’s office where she collected all data for this study. She is a PhD candidate at the Graduate Center—City University of New York, studying Quantitative Methods in Educational and Psychological Research. At the time of publication, she transitioned into a new position as a research scientist for the New York City Department of Probation. Her areas of interest include race, prosecution, item response theory, and multilevel modeling.

Brian D. Johnson

Brian D. Johnson is an associate professor of Criminology and Criminal Justice at the University of Maryland. His areas of expertise involve social inequality in the justice system, with a particular focus on racial disparities in criminal case processing and sentencing. Much of his research examines contextual influences in punishment, as well as the use of advanced statistical modeling techniques to study the criminal process. He is the recipient of the 2008 ASC Ruth Shonle Cavan Young Scholar Award and the 2011 American Society of Criminology, Division on Corrections and Sentencing New Scholar Award.