Take home message
When surveyed, most defenders and prosecutors implicitly calculate an acceptable plea by looking at two factors: the likely sentence at trial and the probability of a guilty verdict.
Research has verified that pleading guilty results in discounted sentences. To date, criminologists have examined what makes these discounts more or less favorable to the defendant, and have found differences depend on the facts of the case, the jurisdiction, and the characteristics of the defendant (including their demographics). These researchers took a different approach, and decided to examine a common assumption about plea bargaining – namely, that a bargain’s acceptability is a function primarily of the sentence a person could expect to receive at trial discounted by the probability they will be acquitted. They call this the ‘shadow of trial’ model.
They designed an elaborate simulation. Defenders, prosecutors and judges were asked to take an online survey which required them to open virtual ‘files’ containing information about an hypothetical robbery case. The files contained information such as the defendant’s age, prior record, and whether or not they had confessed. The information the respondents viewed was tracked. Then they were asked what they thought the likely sentence at trial would be, as well as the probability the defendant would be found guilty and what an acceptable plea would be.
The authors used those data to test the ‘shadow’ model, and found it fit relatively well, although most respondents tended to nominate plea levels slightly lower than the mathematical model suggested. For defenders, this tendency was magnified at higher sentencing levels. Judges were an exception and tended not to be sensitive to the probability of conviction in the way the attorneys were, instead giving fixed discounts of 30-40% to any individual who pled. In an intriguing final analysis, the authors also revealed that their data suggested attorneys who were male and relatively inexperienced said they would hold out for more favorable pleas than other attorneys in the sample.
When surveyed, most defenders and prosecutors implicitly calculate an acceptable plea by looking at two factors: the likely sentence at trial and the probability of a guilty verdict.
- Full report: Shawn Bushway, Allison Redlich and Robert Norris, 2014. “An Explicit Test of Plea Bargaining in the ‘Shadow of Trial.’” Criminology, vol. 52/4, pp.723-754.
- Link: http://onlinelibrary.wiley.com/doi/10.1111/1745-9125.12054/suppinfo. If you have trouble accessing the article contact Andrew Davies.
- Sample: 835 defense attorneys, 372 judges and 378 prosecuting attorneys from all 50 states, solicited non-systematically via national and state defender organizations.
Research has verified that pleading guilty results in discounted sentences. To date, criminologists have examined what makes these discounts more or less favorable to the defendant, and have found differences depend on the facts of the case, the jurisdiction, and the characteristics of the defendant (including their demographics). These researchers took a different approach, and decided to examine a common assumption about plea bargaining – namely, that a bargain’s acceptability is a function primarily of the sentence a person could expect to receive at trial discounted by the probability they will be acquitted. They call this the ‘shadow of trial’ model.
They designed an elaborate simulation. Defenders, prosecutors and judges were asked to take an online survey which required them to open virtual ‘files’ containing information about an hypothetical robbery case. The files contained information such as the defendant’s age, prior record, and whether or not they had confessed. The information the respondents viewed was tracked. Then they were asked what they thought the likely sentence at trial would be, as well as the probability the defendant would be found guilty and what an acceptable plea would be.
The authors used those data to test the ‘shadow’ model, and found it fit relatively well, although most respondents tended to nominate plea levels slightly lower than the mathematical model suggested. For defenders, this tendency was magnified at higher sentencing levels. Judges were an exception and tended not to be sensitive to the probability of conviction in the way the attorneys were, instead giving fixed discounts of 30-40% to any individual who pled. In an intriguing final analysis, the authors also revealed that their data suggested attorneys who were male and relatively inexperienced said they would hold out for more favorable pleas than other attorneys in the sample.
Subscribe: