Directing Retribution: On the Political Control of Lower Court Judges

The sentencing decisions of trial judges are constrained by statutory limits imposed by legislatures. At the same time, judges in many states face periodic review, often by the electorate. We develop a model in which the effects of these features of a judge’s political landscape on judicial behavior interact. The model yields several intriguing results: First, if legislators care about the proportionality of punishment, judicial discretion increases with their punitiveness. Second, voters are limited by two factors in their ability to make inferences about judicial preferences based on observed sentences: the extent to which judges are willing to pander to retain office and the range of judicial discretion mandated by the legislature. Finally, legislators can sometimes manipulate judicial discretion to aid sufficiently likeminded voters in their efforts to replace ideologically dissimilar judges. Lower court judges are fundamental players in the day-to-day operation of the criminal justice system. Voters and their elected representatives delegate responsibility to them to make decisions that, at the extreme, help to determine whether convicted felons live or die. It is perhaps not surprising, therefore, that the exercise of judicial discretion is prone to recurrent political controversy. Conservatives complain that judges ignore citizen preferences regarding the appropriate punishment of the guilty and the contours of public morality more generally (Lambro 2005). Progressive critics of judicial elections express fear that judges may knowingly exceed appropriate punishment in order to appear tough on crime to voters (Croly 1995). Concerns about the behavior of judges echo apprehensions about numerous government officials to whom authority is delegated. Bureaucrats might shirk their responsibilities by implementing policy at odds with the preferences of their elected principals or citizens more broadly (Lowi 1979; Epstein and O’Halloran 1994, 2000; Huber and McCarty 2004), overstate costs to pad their budgets (Niskanen 1971), or engage in other forms of renegade action (Davis 1969). The fear of shirking by agents has led both legislators and executives to expend significant resources to identify competent likeminded bureaucrats, place formal limits on their discretion, and subject them to ex post review and potential dismissal. Contemporary political conflict surrounding the control of judicial behavior echoes that surrounding the control of bureaucrats. Liberals and conservatives alike actively seek to ensure the selection of ideologically similar judges, with important consequences for the composition of the judiciary. Likewise, the push to implement mandatory minimum and maximum sentences, as well as the more general effort to structure judicial sentencing through the imposition of sentencing guidelines and enhancements, are appropriately viewed as attempts by elected officials to prevent judges from assigning sentences at odds with legislative preferences. Finally, the contentiousness of both judicial reappointment proceedings and elections involving incumbent judges underscore persistent concern about the behavior of sitting judges. Political control of trial judges’ authority to sanction punishments has clear parallels with similar efforts to constrain bureaucrats. At the same time, two features of this authority and its context make the application of existing models of delegation in politics problematic. First, political conflict surrounding the nature of punishment at base concerns disagreements over proportionality. While all might concede that a defendant who is not culpable should not be punished, reasonable people disagree about how rapidly punishment should increase with culpability. As we detail below, the nature of these disagreements cannot be adequately captured by canonical spatial models of delegation in politics. Second, judges in 39 states must stand for periodic review and potential replacement by voters. In eight additional states, incumbent judges are reviewed by either commission or other elected officials. Judges therefore differ from bureaucrats, most of whom are protected from replacement by civil service rules. As such, lower court judges are simultaneously constrained ex ante and subject to review ex post. In this article, we incorporate these features into a comprehensive formal model of judicial sentencing discretion. In the model, a legislative body enacts formal constraints on judicial discretion, judges sentence given their own private information regarding their preferences and defendant culpability, and voters decide whether to retain judges given observed sentences. The fact that the actors in the model value proportionality creates important asymmetries between lenient and punitive legislators and voters in their ability to control judicial behavior and replace unlike-minded judges. Surprisingly, it is the most lenient legislators that wish to place the strongest constraints on judicial sentencing. However, more punitive legislators will be generically more displeased with the judicial system than their more lenient counterparts. When judges are constrained ex ante and reviewed ex post, a rich set of strategic interactions exist between the legislator and the voter. The relationship between ex ante constraints and ex post review is complex, so we break the full model into its component parts before considering their interaction. After discussing the model’s assumptions, we first derive optimal ex ante minimum and maximum sentences in the absence of ex post review. Next, we consider ex post voter oversight of judges in the absence of binding ex ante constraints. Finally, we consider ex ante constraints and ex post review in tandem. The model suggests that the ability of voters to make inferences about judicial preferences will be altered by the severity of legislated constraints on judicial discretion and judges’ incentives to pander to voter tastes in order to retain office. In our model, legislators will take these issues into account when deciding how stringent those constraints should be. In some cases, legislators may expand judicial discretion to exploit the ability of likeminded voters to screen ideologically divergent judges. We conclude by discussing several substantive extensions of our framework, the consonance of our model with features of current debates over judicial discretion, and a number of predictions amenable to empirical testing.