UM Study: Race, Geography Factors in Md. Death Penalty Decisions

COLLEGE PARK, Md. Race affects the way death penalty cases are handled in Maryland, mainly influencing prosecutors' decisions early in the process, says a new study from a University of Maryland criminologist and statistician. The study also finds substantial variations in the way Maryland jurisdictions deal with capital cases.

"Disparities in treatment start at the earliest stages of prosecution," says Raymond Paternoster, the study's principal investigator. "These differences are not exacerbated at trial or actual sentencing phases, but they also don't get corrected then. So disparities persist."

Conducted over two-and-a-half years, this statistical analysis of thousands of public records is the most exhaustive study ever of Maryland's application of the death penalty. The researchers examined records of every homicide prosecution in which the death penalty might have been applied between 1978, when the law took effect, and 1999. They found:

• By itself, the offender's race does not play a clear role in the way cases are handled, but the victim's race does make a difference. Defendants accused of killing white victims are significantly more likely to face the death penalty than cases with non-white victims.

  ---

  A copy of the executive summary of the report is available at http://www.newsdesk.umd.edu/pdf/exec.pdf'
  Final report: http://www.newsdesk.umd.edu/pdf/finalrep.pdf

  ---

• When the race of the offender and victim are examined together, the study finds: Black offenders who kill blacks are significantly less likely to face the death penalty, while black offenders who kill whites are significantly more likely to face a death sentence than all other racial combinations.

• Prosecutors in different jurisdictions exhibit considerable variation in the extent to which they seek the death penalty.

• These trends are detected in the early stages of prosecution when state's attorneys decide whether to seek a death sentence. They are not detected later ¿ after conviction, when prosecutors make a final decision whether to pursue a death sentence, or in the final judgment by judges or juries. But the initial disparity is not corrected at these later stages, and so its effects persist.

Methods

Since 1987, four studies have examined charges of racial disparity and arbitrary jurisdictional differences in the handling of Maryland death penalty cases. All have suffered from two deficiencies: They did not look at every case in which the death penalty could have been sought, and they failed to consider the effects of various case characteristics that might masquerade as racial effects and lead to false conclusions.

This study was designed to correct these inadequacies. Based on prison and prosecution records, researchers concluded there were nearly 6,000 homicide prosecutions in Maryland between 1978 and 1999. Then they determined which of these cases were death eligible - met the minimum legal requirements for application of the death penalty. In all but 300 cases, this determination was clear-cut. To handle the ambiguous cases, researchers convened a panel of expert defense and prosecution lawyers. All together, 1,311 case99 met the definition of death eligible.

Researchers then combed the prison, court, prosecution, defense, police and public health records of these 1,311 cases looking for detailed information about characteristics of the defendants, victims and crimes. This yielded about 100 pages of details on each case, which researchers translated into data on 123 possible explanatory case factors.

In their analysis, researchers examined how these case factors correlated with the way cases were handled at four critical decision points - 1) prosecutors' initial decision on whether to seek a death sentence, by filing a notice of intent; 2) subsequent decisions on whether to retract this notice or let it "stick;" 3) prosecutors' decisions to press for a death sentence after conviction; 4) judges' or juries' actual sentences.

Paternoster analyzed the data in two stages. When he found an apparent race and geographic effect, as most earlier studies had, he then statistically "controlled" for the 123 variables - to see whether the effects were real. Could other case characteristics explain what at first appeared to be the effects of race and jurisdiction?

"Statistically, I threw the kitchen sink at these initial findings," Paternoster says. "My hunch was other case characteristics could at least knock out the race effect. But it didn't. The race and geography effects are very robust."

Specific Findings (Final Analysis)

• Maryland state's attorneys are significantly more likely to file initially for a death sentence and make it "stick" (not withdraw it prior to trial) when the race of the victim is white. The race of the victim does not matter later in the process ¿ the decision to pursue a death sentence following conviction and the final judgment by judge or jury. Nevertheless, the effects of the state's attorney's initial decisions are passed on to these later stages. By itself, the race of the defendant does not have any clear effect on decisions at any stage.

• Maryland state's attorneys are significantly less likely to file initially for a death sentence when the homicide is a black-on-black killing. This effect is not manifest at any of the other three stages of the process. But the initial disparate treatment is not corrected later, and so is passed on to later points in the process.

• Maryland state's attorneys are significantly and substantially more likely to file initially for a death sentence when a killing involves a black offender and white victim compared with other homicides. This effect is not manifest at any of the other three stages of the process. But the disparate treatment at the initial stage is not corrected later, and so is passed on to later points in the process.

• Baltimore County is significantly more likely to file initially for a death sentence than other Maryland jurisdictions. Later in the process, Anne Arundel and Baltimore Counties are significantly more likely to make it "stick" compared to other jurisdictions. No county-to-county differences are manifest later in the process. Nevertheless, the early differences have effects that persist at all stages.

03001