Politics, Competence, and Performance: Evidence from the US State Budget Agencies
| Published date | 01 May 2021 |
| Author | Jinhai Yu,Edward T. Jennings |
| Date | 01 May 2021 |
| DOI | http://doi.org/10.1111/puar.13277 |
Research Article
500 Public Administration Review • May | June 202 1
Abstract: The relationship between administrative competence and political responsiveness is central to public
administration theories. This study examines the conditional effects of competence on performance as moderated by
politics. Synthesizing the theories of neutral competence and responsive competence, we propose a model of contingent
competence. We argue that the impact of competence on performance is conditional on both the preferences of political
principals and the degree of political control as indicated by agency politicization. With multiple indicators of state
fiscal performance, the empirical test uses data on US state budget agencies between 1986 and 2008. The results
support the proposition. When budget agencies are highly politicized, competence of budget agencies influences state
fiscal performance in accordance with the preferences of Democratic and Republican elected officials when one or the
other party controls state governments. The findings enhance the understanding of how institutional designs shape
bureaucratic politics and agency performance.
Evidence for Practice
• Agency competence matters to government performance, as shown in the case of US state budget agencies.
• The impact of agency competence on government performance depends on both policy preferences of
elected officials and the level of political control.
• Agency competence has a greater impact on performance when a high level of agency politicization allows
two-way interactions between elected officials and administrative staffs.
Politics, Competence, and Performance:
Evidence from the US State Budget
Agencies
The relationship between politics and
administration is central to the theoretical
debates of American public administration
(Moynihan and Ingraham2010; Svara2001,
2008). The well-known doctrine of the “politics–
administration dichotomy” holds that politics and
administration should be separated (Wilson1887);
public administrators are expected to maintain neutral
competence. Thus, the more competence is insulated
from politics, the greater the impact competence has
on performance (Krause and Douglas2005). The
“politics–administration dichotomy” was questioned
subsequently (Hollibaugh Jr2014). Opponents argue
that politics and administration are intertwined;
public administrators are expected to maintain
responsive competence (Moe1985). It follows that,
the more closely competence is integrated with
politics, the greater its impact on performance. The
two models of neutral competence and responsive
competence make contradictory predictions on the
impact of competence on performance as moderated
by politics.
We examine the impact of politics and competence
on performance by introducing the direction and the
degree of political control of administrative agencies.
Responsive competence implies that competence can
be used as a tool to achieve policy goals set by political
principals. But who are the political principals? How
much political control is possible? These questions
remain unanswered. There can be multiple political
principals. The degree of political control can vary
across administrative agencies. As Aberbach and
Rockman(1988, 609) argue, “the real issue often is
not politics versus neutral competence but clarifying
the principals (and their underlying principles) in the
principal–agent relationship.”
To answer this call, we propose a third model of
contingent competence by arguing that competence
can have differential impacts on performance
depending on the policy preferences of political
principals and the politicization level of administrative
agencies. We use the US state budget agencies as an
empirical setting to test the contingent competence
model. Budgeting has been long recognized as
both political and technical (Thurmaier and
Willoughby2002; Wildavsky1964), where politics
Jinhai Yu
Shanghai University of Finance and Economics
Politics, Competence, and Performance: Evidence from the
US State Budget Agencies
Edward T. Jennings Jr.
University of Kentucky
Edward T. Jennings, Jr. is professor
emeritus and provost’s distinguished service
professor of Public Policy and Administration
at the Martin School of Public Policy and
Administration, University of Kentucky,
USA. His primary research and teaching
interests include policy formulation and
implementation and public management.
Email: ed.jennings@uky.edu.
Jinhai Yu is assistant professor at
the School of Public Economics and
Administration, Shanghai University of
Finance and Economics, China. He teaches
public policy analysis and public budgeting.
His research focuses on public budgeting
in the nexus of politics and public
administration.
Email: yu.jinhai@mail.shufe.edu.cn
Public Administration Review,
Vol. 81, Iss. 2, pp. 500–518. © 2020 by
The American Society for Public Administration.
DOI: 10.1111/puar.13277.
Politics, Competence, and Performance: Evidence from the US State Budget Agencies 501
and competence can influence performance. To measure fiscal
performance, we use four indicators of financial condition (Wang,
Dennis, and Yuan Sen2007) and credit ratings (e.g., Lewis2012).
We find that the impact of budget agency competence on state
fiscal performance is conditional on government partisan control
and the level of politicization. When the budget agencies are highly
politicized, budget agency competence increases general fund
balance and credit ratings when the Republican Party has unified
control of state governments; it decreases budget balance and credit
ratings when the Democratic Party holds unified government
control. Furthermore, the competence of highly politicized
budget agencies shows a larger, positive effect on debt level and
total expenditure under Democratic than Republican control.
The empirical findings provide limited support for the neutral
competence model and revise the responsive competence model.
They support a model of contingent competence where politics is
framed in two dimensions—both the direction and the degree of
political control of administrative agents.
The findings suggest that budget agency competence is effective in
shaping government fiscal performance. To understand the exact
impact of the competence, however, one needs to account for
the policy preferences of elected officials and the level of agency
politicization. The policy preferences can determine whether
budget agency competence contributes to conservative or liberal
fiscal performance, as preferred by the Republican Party and
the Democratic Party, respectively. The policy preferences have
larger impacts as agency politicization increases because it allows
stronger political control and more communication between elected
officials and administrative staffs. This echoes Thurmaier and
Willoughby’s(2002) finding that state budget staffs show greater
policy influences when they feel a shorter “policy distance” and get
well informed of governors’ policy priorities.
This study makes three contributions. First, it adds to the literature
on the effects of administrative competence on government
performance. It shows that competence and politics combine
to influence performance. This refutes the conventional view
of a politics–administration dichotomy and enhances our
understanding of the effectiveness of public administration in
general. Second, it contributes to the literature on political control
of administrative agencies. It shows that one may better understand
the role of political control by distinguishing the preferences of
political principals and the degree of control that they have over
administrative agents. Third, the empirical tests contribute to the
fiscal performance literature by demonstrating the critical roles
played by budget agencies and the importance of the interactive
effects between politics and management factors.
Previous Studies
Originated from the doctrine of the politics–administration
dichotomy, neutral competence is the conventional model on
the relationship between administrative competence and political
responsiveness. Neutral competence indicates that administrative
competence should be impartial or nonpartisan (Kaufman1956).
Empirical studies generally show that administrative competence
improves organizational performance (Carpenter2001). For
instance, Krause and Douglas(2005) find that agency performance
is shaped by reputational considerations but not political influences,
a finding consistent with neutral competence. Schelker(2012b)
finds that auditor expertise as indicated by a Certified Public
Accountant (CPA) requirement is associated with higher credit
ratings, a proxy of government fiscal performance. Nevertheless,
Demir and Nyhan(2008) find no empirical support for the politics–
administration dichotomy in council-manager local governments.
An alternative to neutral competence, responsive competence implies
that political responsiveness should be prioritized over administrative
competence (Moe1985). Responsive competence is mainly achieved
via politicization. Many empirical studies find that politicization
decreases organizational performance. Lewis(2007) and Gallo
and Lewis(2012) find that federal agencies headed by political
appointees show lower performance as measured by scores under the
Program Assessment Rating Tool (PART). Wood and Lewis(2017)
find that politicized federal agencies respond more slowly to public
information requests, indicating poor management performance.
Lapuente and Suzuki(2020) find that the politicization of
bureaucracies is associated with lower pro-innovation attitudes.
Nonetheless, Mikesell and Ross(2014) find that the politicization of
state revenue forecasts contributes to forecast acceptance, implying
one positive performance impact of politicization.
Scholars have identified a tradeoff between administrative
competence and political responsiveness and argue for a balance
as the recipe for better organizational performance (Moe1985).
Studies have shown that agency politicization crowds out neutral
competence (Christiansen, Niklasson, and Öhberg2016; Öhberg,
Munk Christiansen, and Niklasson2017) and policy expertise
(Richardson2019). In addition, Krause, Lewis, and Douglas(2006)
propose a balance between the use of political appointments and
civil service systems in the recruitment of budget directors and
budget staff to maximize budget agency performance. They find
empirical support for this proposition. Bourdeaux(2008) finds a
tradeoff between politics and neutral competence in two cases of
solid waste policies of New York State. She further identifies a case
where a balance between politics and competence seems practical.
In the case of state budget agencies, both administrative competence
and political responsiveness seem to matter. Thurmaier and
Willoughby(2002) characterize state budget offices on a continuum
from control orientation to policy orientation. In control-oriented
budget offices, budget staff find it easier to maintain neutral
competence. In policy-oriented budget offices, budget staff strive
to balance neutral competence with political responsiveness to
governors. In addition, Thurmaier and Willoughby(2002) find that
budget staff have greater policy influence when maintaining a closer
relationship with governors. A shorter “policy distance” (Thurmaier
and Willoughby2002) helps the budget staff keep abreast of
governors’ policy priorities and increases the level of governors’
trust in the budget staff. To the extent that a shorter policy distance
indicates stronger gubernatorial control of budget agencies, this
suggests that politicization may increase the impact of state budget
agencies on government fiscal performance.
We examine the impact of competence of budget agencies on
state fiscal performance, extending the studies on the performance
impact of administrative competence. Building upon Thurmaier
To continue reading
Request your trial