Cues of Deservingness and Street-Level Decision Making: Evidence from a Conjoint Experiment among US teachers

Street-level bureaucrats have a great deal of freedom in bureaucratic decision-making, ranging from whether they grant clients a claim, provide them with additional information, or simply offer some extra help. While frontline... [ view full abstract ]

Street-level bureaucrats have a great deal of freedom in bureaucratic decision-making, ranging from whether they grant clients a claim, provide them with additional information, or simply offer some extra help. While frontline workers in their daily interactions with citizens ought to exercise bureaucratic impartiality, a large literature highlights potential discretionary biases in street-level decision-making. Here, it is argued that street-level bureaucrats with high workloads and low financial resources at work would be more prone to focus on those clients they deem deserving (or worthy). While numerous studies have empirically shown that frontline workers indeed base their decisions on individual judgements of clients, we know relatively little about what determines those judgements. In one of the rare experimental applications on street-level decision-making, Scott (1997) shows that evoked compassion leads frontline workers to become more likely to grant clients benefits, but argues that other factors such as race or social class may signal similar cues of deservingness. Indeed, street-level workers actively cope with conflicting demands on their time and labor by prioritizing certain types of clients. But how do street-level bureaucrats decide which clients deserve their time and labor?

In this study, we outline a behavioural model of how client characteristics affect frontline workers decision-making, and test it empirically. Doing so, we focus on an important policy area – education -- where there exist only few empirical micro-level applications. Indeed, studying teachers’ judgements and decision-making lies at the core of bureaucratic discretion, and US educators face considerable discretion in their decisions of whether or not to provide additional help to their students, for example, after class. Not only can educators decide if they invest extra time and effort in their students, they typically have to choose with whom they do so. This can lead to discriminatory practices -- favouring a particular group of students over another. Especially in times of limited resources, this is not only a problem of theoretical concern, but one of great practical pedigree for the social equity of the US educational system.

We experimentally test the effectiveness of information cues of students’ personal characteristics on street-level bureaucrats’ stated helping behavior among a nationwide sample of US educators. We independently randomize student characteristics which signal deservingness within a conjoint analytical specification by presenting student profiles of multiple pairs to US educators and subsequently asking them each time to select one of the profiles they would help.