Bureaucratic Beliefs and Representation: Linking Social Identities, Attitudes, and Client Outcomes

• Published date: 01 May 2024
• DOI: http://doi.org/10.1177/02750740231213995
• Author: Nathan Favero
• Date: 01 May 2024
• Subject Matter: Articles

Bureaucratic Beliefs and Representation:

Linking Social Identities, Attitudes, and

Client Outcomes

Nathan Favero

Abstract

Representative bureaucracy theory posits that the demographic makeup of a bureaucracy can affect how policy is imple-

mented, especially when bureaucrats engage in “active representation”or behavior that directly advances the interests of

a particular group in society. It is often assumed that active representation is motivated by the unique beliefs, convictions,

or aff‌inities experienced by bureaucrats holding particular social identities. But few studies of representative bureaucracy

have attempted to directly measure the attitudes of bureaucrats, and even fewer studies examine whether such attitudes

are meaningfully linked to policy outcomes. This study examines the social identities, self-perceived roles, and political pref-

erences of local school administrators in Texas. The results conf‌irm a link between bureaucratic managers’social identities

and distributional policy outcomes, while also suggesting that distinctive bureaucratic attitudes (as obser ved here) can

offer at best a partial explanation for why the social identities of bureaucrats are linked to policy outcomes.

Keywords

representation, bureaucracy, race, policy implementation, schools

Representation is central to many conceptions of governance.

While elected off‌icials offer the most obvious route for the

representation of public interests to occur within a democ-

racy, the concept of representation has been extended to

some policy decision making processes among unelected

off‌icials, independent of any direction they might receive

from elected politicians. For example, participatory budget-

ing can directly engage members of the public in government

decision making in a way that affords representation to

various public interests (Barbera et al., 2016; Rossmann &

Shanahan, 2012). And the theory of representative bureauc-

racy—the focus of this study—posits that if members of a

particular social group are found among the employees of a

bureaucracy, there may be some meaningful representation

afforded to that social group by the bureaucracy. While

early scholarship on representative bureaucracy was largely

normative, the theory has inspired a large body of empirical

work, much of it supporting the notion that clients of a gov-

ernment program tend to enjoy greater benef‌its when they

encounter bureaucrats who share a salient demographic trait

with them (see Bishu & Kennedy, 2020; Kennedy, 2014).

Many studies of this linkage have used observational

research designs that leave them potentially vulnerable to

omitted variable bias, but recent quasi-experimental work

in the f‌ield of education has helped conf‌irm a causal link

between racial matching of client-to-bureaucrat and positive

client outcomes (Gershenson et al., 2022; Redding, 2019).

Despite such advances to our knowledge of this linkage,

we have a limited understanding of the mechanisms by

which such relationships typically function, although many

possibilities have been identif‌ied (Lim, 2006; Nicholson-

Crotty et al., 2016).

Traditionally, one of the key mechanisms of bureaucratic

representation that scholars have pointed to is bureaucratic

attitudes. Yet few studies of representative bureaucracy actu-

ally attempt to measure attitudes, especially in recent

decades. While representative bureaucracy scholarship has

largely ignored the measurement of beliefs or preferences,

an expansive literature on public service motivation has

made surveying bureaucrats about their attitudes a central

feature of the public administration literature. The public

service motivation literature has drawn attention to variation

among bureaucrats in how motivated they are to serve the

public interest, but it has told us much less about potential

conf‌lict among bureaucrats in their visions of what it

Department of Public Administration & Policy, American University,

Washington, DC, USA

Corresponding Author:

Nathan Favero, American University, Department of Public Administration

& Policy, School of Public Affairs, 4400 Massachusetts Ave. NW,

Washington, DC 20016, USA.

Email: favero@american.edu

Article

American Review of Public Administration

2024, Vol. 54(4) 337–353

© The Author(s) 2023

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/02750740231213995

journals.sagepub.com/home/arp

means to serve the public interest. Such conf‌licts may be

especially likely to arise when different segments of the

public have competing interests, as is often the case when

implementing important public policies or administering

key government programs.

This study examines bureaucratic attitudes about repre-

sentation and politics in the context of K-12 public schooling.

Many institutional structures for public schooling in the U.S.

were traditionally designed with an eye toward sheltering

them from partisan political conf‌licts (Lay & Bauman,

2019; Tyack, 1980). Nonetheless, we see highly contentious

debates over school policy frequently emerging in American

politics (Alvord & Rauscher, 2021; Boyd, 2007; Johnson,

2022; Meier & Rutherford, 2017), suggesting that schools

are a setting ripe for an examination of how representation

plays out practically in the administration of a key govern-

ment service.

I conduct my empirical analysis using a unique dataset of

Texas schools which allows me to pair principal survey

responses from over 500 schools with a variety of archival

measures of school characteristics and outcomes. I f‌ind that

Black and Hispanic principals tend to differ on average

from White principals in their political ideology and their

self-perceived role in representing minoritized student inter-

ests. Still, there is substantial overlap in viewpoints among

principals of different ethnoracial identities. I also f‌ind that

principal race/ethnicity is signif‌icantly associated with

several measures of ethnoracial inequity in school inputs

and outputs. Fewer signif‌icant relationships are found for

principal attitudes, but I do f‌ind evidence that attitudes are

associated with some student outcomes.

Employee Attitudes and Representative

Bureaucracy Theory

Studies of representation in the bureaucracy have identif‌ied

several mechanisms by which the demographic makeup of

a government organization might be linked to policy out-

comes. Studies often point to socialization experiences that

are connected to social identities, resulting in values, priori-

ties, or aff‌inities being associated with demographic traits

among bureaucrats (Fernandez et al., 2018; Meier, 1993b;

Vinopal, 2020; see also Sampson, 2019). Additionally,

certain bureaucratic skills, such as language f‌luency, that

help serve particular segments of society may be associated

with particular demographic traits (Fernandez et al., 2018;

Meier, 2019). Bureaucrats’attitudes and abilities can then

affect bureaucratic behavior, ultimately inf‌luencing organiza-

tional outcomes. A bureaucrat’s demographic traits can also

affect outcomes more indirectly, as when members of the

public react favorably to seeing a bureaucrat who looks

like them or when other bureaucrats change their behavior

in response to their peers (Li, 2021; Nicholson-Crotty

et al., 2016; Riccucci et al., 2016). It is also possible that

the demographic makeup of an organization and its policy

outcomes are correlated with one another in part because

they are both directly affected by managerial priorities or

other institutional factors (see Favero & Molina, 2018).

Despite the myriad possible mechanisms of representation

identif‌ied in recent scholarship, bureaucratic policy prefer-

ences and related attitudes were emphasized in much of the

early work on the topic. This is perhaps due in part to the

early emphasis in this literature on normative questions

regarding bureaucratic accountability and interest in the pos-

sibility that demographic diversity brings with it a diversity

of values or public interests to the bureaucratic decision

making process (Long, 1952). Yet the strength of the associ-

ation between demographic characteristics and attitudes has

also been questioned, most notably by Meier and Nigro

(1976) who found that most demographic characteristics

are poor predictors of bureaucrats’political attitudes. Of

the factors they examined, race appeared to be the most

important predictor of attitudes. Meier (1993b) later argued

that race is the demographic characteristic most likely to

produce linkages between passive and active representation

in the United States because so many political issues in the

United States are racialized. It is perhaps no coincidence

that race has been the demographic characteristics most

often studied within the representative bureaucracy literature,

with gender receiving the second largest share of attention

(Bishu & Kennedy, 2020; Kennedy, 2014).

Despite the longstanding assumption that values or atti-

tudes provide a mechanism by which the demographic

makeup of bureaucrats affects both bureaucratic behavior

and substantive outcomes, very few studies directly

examine the attitudes of bureaucrats (Lim, 2006). Selden

(1997) provides perhaps the only quantitative study of repre-

sentative bureaucracy that directly models the mediating

effect of attitudes in the relationship between bureaucrats’

demographic characteristics and policy outcomes. In her

examination of rural loan programs, she used a survey to

measure the extent to which supervisors saw themselves as

representatives of minority interests within their workplace

—a construct she called minority representative role accep-

tance. She found that supervisors who saw themselves as rep-

resentatives of minority interests approved more loans for

minoritized applicants. Minoritized supervisors tended to

report higher levels of representative role acceptance than

White supervisors, but it was ultimately supervisors’

responses to this survey measure—not their race—that inde-

pendently predicted the rate at which they approved loans

from minoritized applicants. When it came to publicizing

loan programs, however, both the representative role accep-

tance measure and the supervisor’s race were signif‌icant

independent predictors of the self-reported efforts to reach

out to minoritized communities. More recently, drawing on

data from a survey study conducting in connection with a

local government, Bradbury and Kellough (2008) found

that compared to White government administrators, Black

338 American Review of Public Administration 54(4)