Racial Discrimination and Street‐Level Managers: Performance, Publicness, and Group Bias

• Published date: 01 November 2021
• Author: Zachary W. Oberfield,Matthew B. Incantalupo
• Date: 01 November 2021
• DOI: http://doi.org/10.1111/puar.13376

Research Article

Racial Discrimination and Street-Level Managers: Performance, Publicness, and Group Bias 1055

Abstract: This article broadens our understanding of street-level governance by examining how citizen performance,

organizational publicness, and group bias moderate racial discrimination among street-level managers (SLMs). We

examine this topic with an experiment in which we requested enrollment information from public and charter school

principals while randomly assigning a putative student’s race and ability. As expected, SLMs discriminated based

on race, and positive performance information mitigated this discrimination. Surprisingly, negative performance

information also reduced discrimination. Turning to publicness, we find no evidence that less public organizations

(charter schools) exacerbated anti-Black discrimination. Finally, we show that White SLMs discriminated against

Black citizens. However, Black SLMs worked in more administratively difficult settings and, perhaps as a result,

responded at lower rates; thus, Black citizens were equally likely to receive responses from White and Black SLMs.

Therefore, improving access to public agencies may require representativeness and support for SLMs working in

challenging organizational environments.

Evidence for Practice

• Antiracist efforts need to start at the top: Ensuring equal access to public services requires reckoning with

bias among street-level managers.

• When public servants get more individualized information about citizens, they may be less likely to default

to traditional group stereotypes.

• Access to public services may be conditioned by the interplay of citizens’ and bureaucrats’ identities. Public

organizations should strategize about concrete steps they can take to reduce group bias, like implicit bias

training and antiracist hiring practices. In addition, when minority administrators are hired to difficult posts,

they need to be supported so that they can better serve disadvantaged populations.

• Using less public forms of organization to deliver public services may not increase racial discrimination.

Recent audit studies reveal disconcerting

evidence of racial and ethnic discrimination

in public agencies (Einstein and Glick 2017;

Giulietti, Tonin, and Vlassopoulos 2019; Jilke, Van

Dooren, and Rys 2018; Olsen, Kyhse-Andersen, and

Moynihan 2020; White, Nathan, and Faller 2015).

This article contributes to our understanding of

this problem in two ways. First, to date, most

studies in this literature have focused on street-level

bureaucrats (SLBs), the front-line employees who

meet directly with citizens1 and implement public

policy (Lipsky 1980). Although SLBs play a crucial

role in governing access to public services, we know

little about racial discrimination among street-

level managers (SLMs), the leaders or executives of

street-level organizations (Gassner and Gofen 2018).

Scholars have traditionally portrayed SLMs as

marginal to street-level governance (Lipsky 1980);

however, recent works indicate that they play an

underappreciated role in determining citizen access to

public services. As such, we study racial discrimination

among SLMs.

Second, although we have evidence of racial

discrimination in public agencies, few studies have

asked about the factors that moderate discrimination.

Therefore, we have little understanding of what can

be done to mitigate this problem. In response, this

article examines three potential moderators of racial

discrimination: citizen performance, organizational

publicness, and group bias. By studying whether these

factors exacerbate or mitigate racial discrimination,

this article improves our understanding of street-

level organizations and identifies practical steps

organizations can take to reduce biased behavior

(Nagtegaal et al. 2020).

To make these contributions, this article presents

findings from an audit study of a nationally

representative sample of U.S. public high school

Racial Discrimination and Street-Level Managers:

Performance, Publicness, and Group Bias

Zachary W. Oberfield Matthew B. Incantalupo

Haverford College Yeshiva University

Matthew B. Incantalupo is an

Assistant Professor of Political Science at

Yeshiva University. His research interests

include political behavior, inequality, and

experimental methods. He is a coauthor

of

Elite-led Mobilization and Gay Rights:

Dispelling the Myth of Mass Opinion

Backlash

, University of Michigan Press.

Email: matthew.incantalupo@yu.edu

Zachary W. Oberfield is a Professor

of Political Science at Haverford College.

His research interests include education,

leadership, and street-level bureaucracy.

He is the author of two books:

Are

Charters Different? Public Education,

Teachers, and the Charter School Debate

,

Harvard Education Press, and

Becoming

Bureaucrats: Socialization at the Front

Lines of Government Service

, University of

Pennsylvania Press.

Email: zoberfie@haverford.edu

Public Administration Review,

Vol. 81, Iss. 6, pp. 1055–1070. © 2021 by

The American Society for Public Administration.

DOI: 10.1111/puar.13376.

1056 Public Administration Review • November | Decemb er 2021

principals. In the experiment, we requested enrollment information

while randomly assigning the ability (low ability, high ability, or no

mention of ability) and race (Black or White) of a putative student.

Our outcome of interest is whether the principal responded to the

inquiry within 30 days.

Public schools are generally required by law to serve all students

residing in their geographic districts. Nevertheless, there is a

variety of practices schools can enact to impede student access

(Ravitch 2013). In addition, school officials may hinder access

by unconsciously discriminating between citizens (Fiske 1998).

Whatever the reason, failing to respond to a request for enrollment

information imposes an administrative burden on citizens,

who must seek out such information elsewhere (Herd and

Moynihan 2018). These burdens are particularly onerous for

members of marginalized communities who have fewer resources

to overcome them. In addition, nonresponsiveness may indicate to

citizens that government organizations do not work and that they

are not valued members of the polity (Soss 1999).

To examine how publicness moderates racial discrimination, our

sample includes public district schools (hereafter “district schools”),

which are publicly funded and operated, and public charter schools

(hereafter “charter schools”), which are publicly funded and

privately operated. Charter schools are theorized as “less public”

relative to district schools (Oberfield 2017). As some research

suggests that less public organizations discriminate more (Bergman

and McFarlin 2018; Jilke, Van Dooren, and Rys 2018), comparing

district and charter schools enables us to examine how publicness

is related to racial discrimination. By collecting demographic

information about principals, we observe whether group bias

moderated racial discrimination; put differently, we examine

whether Black and White principals responded at similar rates to

the Black and White student treatments.

Our article proceeds as follows. First, we consider the role that

SLMs play in determining access to public services. Second, we

discuss why SLMs may discriminate based on race and review the

racial discrimination audit literature. Third, we argue that citizen

performance, organizational publicness, and group bias may

moderate racial discrimination. Fourth, we discuss the details of

our experiment and our approach to analyzing our data. Fifth, we

present and discuss our findings.

Our results reveal that principals responded at lower rates to

Black students and that our ability cues, which we interpret as

performance treatments, moderated anti-Black discrimination.

In the absence of ability information, principals discriminated

against Black students; when we indicated that Black students were

high ability, this effect disappeared. Surprisingly, the low-ability

cue also reduced anti-Black discrimination: Even when principals

were presented with ostensibly negative performance information,

they were less likely to discriminate against Black students. Our

examination of publicness also reveals different results than we

expected. We show no evidence of anti-Black discrimination

in charter schools; rather, although not at the level of statistical

significance (p = .054), we reveal suggestive evidence of higher levels

of anti-Black discrimination in district schools. However, because

district school principals responded at somewhat higher rates than

charter school principals, Black students were equally likely to

receive a response from charter and district principals.

Finally, we show that White principals were less likely to respond

to Black students. This suggests that group bias is a significant

moderator of racial discrimination. However, Black principals

worked in more administratively difficult settings and, perhaps as a

result, responded at lower rates on average. As such, Black students

were equally likely to receive a response from White and Black

principals. This complicates the connection between representative

bureaucracy and access to public service and highlights the

importance of supporting administrators assigned to difficult posts.

We conclude the article by considering the practical and theoretical

implications of these findings.

SLMs and Access to Public Services

Since Lipsky’s (1980) pathbreaking work, most street-level bureaucracy

studies have examined the agents at the front lines of government,

SLBs, rather than their superiors, SLMs. This trend was influenced

by implementation studies in the 1970s, which saw official policy

as disconnected from front-line practice and a growing desire to

understand public services from the “bottom-up” (Pressman and

Wildavsky 1973). SLMs also fell from view because Lipsky (1980, 50)

portrayed them as distant and relatively constrained in shaping front-

line practice. In large part, scholars that followed Lipsky reflected and

amplified this view (Brehm and Gates 1997; Brodkin 1997; Maynard-

Moody and Musheno 2003; Oberfield 2014; Riccucci 2005).

Recently, however, studies have begun to question this perspective,

suggesting that street-level supervisors and leaders play a more

important role in the provision of public services than those

traditionally recognized (Hupe and Keiser 2019). For example,

Gassner and Gofen (2018) argue that SLMs serve several key functions

for street-level organizations, including translating formal policy into

practice, adapting arrangements to solve implementation problems,

and articulating street-level information upward to policy makers.

May and Winter (2009) find that SLBs’ responses to policy reforms

were guided by various forces, including the amount of supervision

and degree of delegation from managers. They interpret their results as

indicating that “management matters,” even in street-level bureaucracy.

Even Lipsky (2010, 212), in the thirtieth anniversary edition of his

book, reflects this evolving consensus, noting that the behavior of

managers influences access to public services.

As scholars have refocused on SLMs, they have noted that they

have a different vantage point and set of incentives relative to SLBs

(Gassner and Gofen 2018). Most importantly, while SLBs are

judged on their individual performance, SLMs are assessed on how

their organizations perform. In other words, SLMs are accountable

for the achievement of organization-wide goals. This difference has

implications for how SLMs behave—for example, they are more

likely than SLBs to be concerned with organizational performance

as a whole—and makes them an important group to study. We

elaborate on the implications of this positioning below when we

consider how SLMs respond to performance pressures.

Exploring SLM Racial Discrimination

If SLMs help determine access to public services and are

incentivized to focus on their organization’s performance, how