A replication of “exploring and explaining contracting out: Patterns among the American states”
| Published date | 01 December 2022 |
| Author | Jing Qian,Jiahuan Lu,Jianzhi Zhao |
| Date | 01 December 2022 |
| DOI | http://doi.org/10.1111/padm.12890 |
REPLICATION
A replication of “exploring and explaining
contracting out: Patterns among the American
states”
Jing Qian
1
| Jiahuan Lu
2
| Jianzhi Zhao
3
1
Department of Politics, Princeton University, Princeton, New Jersey, USA
2
School of Public Affairs and Administration, Rutgers University, Newark, New Jersey, USA
3
School of International Relations and Public Affairs, Fudan University, Shanghai, China
Correspondence
Jianzhi Zhao, Fudan University - School of
International Relations and Public Affairs, 220
Handan RD, Shanghai 200433, China.
Email: jianzhizhao@fudan.edu.cn
Funding information
National Office for Philosophy and Social
Sciences of China, Grant/Award Number:
19ZDA072; National Science Foundation of
China, Grant/Award Number:
2020YFA0908600
Abstract
This manuscript conducts both a narrow and wide replica-
tion of Brudney et al. (2005, Journal of Public Administra-
tion Research and Theory, 15[3], pp. 393–419), focusing on
the scope, perceived effectiveness, and antecedents of con-
tracting out. The replication joins the original study in indi-
cating that state agencies widely used contracting out in
service delivery, but their perceived effects on service costs
and quality were mixed. The replication coincides with the
original study in suggesting that contracting out was an
agency-specific practice driven more by pragmatic factors
but reports some divergences in the effects of individual
factors. Overall, our replication concurs with the original
study in many aspects, but also adds new nuances. There
seems to be no single pattern underlying the use of con-
tracting out, and thus no single factor could determine con-
tracting out under all circumstances. Rather, contracting is a
dynamic practice used by agencies for certain contingencies.
Abstract
本论文对Brudney 等人(2005, Journal of Public Administra-
tion Research and Theory, 15[3], pp. 393–419) 的重要研究
进行了狭义和广义上的结果再现，并重点关注政府外包的范
围、有效性感知和决定性因素。基于更新和更全面的数
据，本文的结果总体与原文保持一致，即尽管州机构在公
共服务中广泛使用外包策略，但它们对公共服务成本和质量
Received: 6 February 2022 Revised: 16 September 2022 Accepted: 1 October 2022
DOI: 10.1111/padm.12890
Public Admin. 2022;100:1161–1182. wileyonlinelibrary.com/journal/padm © 2022 John Wiley & Sons Ltd. 1161
的感知却参差不齐。尽管单个变量的偏效应存在部分差
别，但本文的稳健结果再次证明，外包决策主要取决于个体
机构的特征，更多地受现实需要驱动。此外，我们的论文发
现与原文存在一些细微的差异，我们发现外包决策的决定并
不受单一稳定因素的驱动，相反，它是随着机构的变化而动
态变化的.
1|INTRODUCTION
Since the New Public Managementmovement in the 1980s, contracting out has beenwidely used in various countries
as a public service delivery mechanism (Hood, 1991; Kelman, 2002; Pollitt & Bouckaert, 2004). Instead of relying on
public employees to directly serve citizens (in-house delivery), governments form contractual relationships with third-
party organizations and hire them to deliver publicly funded services to citizens (contracting out). As government con-
tracting becomes a widespread government tool in public service delivery, it attracts substantial scholarly attention.
Within the long volume of literature on government contracting, one well-cited research is Brudney et al. (2005). The
study undertakes an early, systematic examination of government contracting among American states. Using survey
data from stateagency leaders in the 50 Americanstates in 1998 as well as sociodemographic data, the studyexamines
three importantquestions on government contracting, includingits scope, perceived effectiveness, andantecedents.
This manuscript replicates Brudney et al.'s (2005) results in both a narrow and wide sense. First, we narrowly
replicated the study using similar data and model specifications. Second, we replicate the study in a wide sense by
using panel data from later years while sticking to the original specification. Both replications concur with the original
study in many aspects but also add new and interesting nuances. Specifically, both the original study and our replica-
tions consistently show that state agencies across the United States engaged extensively in contracting out in deliv-
ering services to the public, but these agencies typically allocated a modest proportion of their agency budgets to
this practice. Moreover, the original and replication studies similarly report that the state agency leaders' perception
of the effectiveness of contracting out was split, with approximately half of the leaders agreeing that contracting out
improves service quality, while only a third of them suggesting that contracting out contributes to cost savings.
Concerning the antecedents of contracting out, our replication coincides with the original study in several ways.
Overall, contracting out is an agency-specific practice shaped more by individual agency contexts than by the
broader sociodemographic environment. In addition, contracting out is driven more by pragmatic factors than ideo-
logical and political factors.
Specific to the effects of individual antecedents, there are some divergences between the original s tudy and the repli-
cation. Although these differences could result from data and variable measurements, they might imply the dynamics and
complexity of contracting out. In other words, contracting out is a dynamic practice used by agencies to serve pragmatic
needs under certain contexts. There is no single factor that could dominate the explanation of contracting out.
The remainder of the paper is organized as follows. The second section provides a brief synopsis of the original study
being replicated. Then, we discuss our replication methods in the third section, including the data and variables. The fourth
section presents our replication results. We discuss the findings and conclude the manuscript in the fifth section.
2|ASYNOPSISOFORIGINALSTUDY
The significant use of contracting out in public service delivery motivates scholars to examinevarious aspects of gov-
ernment contracting. Brudney et al. (2005) is an early study in this line of inquiry with a focus on the American
1162 QIAN ET AL.
To continue reading
Request your trial