What Explains Agency Heads’ Length of Tenure? Testing Managerial Background, Performance, and Political Environment Effects

• Author: Alice Moseley,Nicolai Petrovsky,George A. Boyne,Oliver James
• DOI: http://doi.org/10.1111/puar.12735
• Published date: 01 July 2017
• Date: 01 July 2017

What Explains Agency Heads’ Length of Tenure? Testing Managerial Background, Performance, and Political Environment Effects 591

Public Administration Review,

Vol. 77, Iss. 4, pp. 591–602. © 2017 by

The American Society for Public Administration.

DOI: 10.1111/puar.12735.

Alice Moseley is lecturer (education and

research) in the Department of Politics at

the University of Exeter, United Kingdom.

Her research is broadly concerned with

understanding citizens’ interactions with

the state, including coproduction and

participation in public services and civic/

political engagement more broadly. She also

has interests in the application of behavioral

economics to public policy. Much of her

recent work makes use of experimental

methods, especially field experiments.

E-mail: a.moseley@exeter.ac.uk

George A. Boyne is professor of public

management at Cardiff University, United

Kingdom. His main research interest is

organizational performance in the public

sector. He is currently working on topics

that include executive succession and

administrative intensity.

E-mail: boyne@cardiff.ac.uk

Oliver James is professor of political

science at the University of Exeter, United

Kingdom. He works on the public policy and

politics of public services, citizen–provider

relationships, public sector organization

and reform, executive politics (particularly

politician–administrator relations), and

regulation of publicly owned and/or funded

bodies and services. He also works on

methodology for using experiments in

public management research.

E-mail: o.james@exeter.ac.uk

Nicolai Petrovsky is associate professor

in the Martin School of Public Policy and

Administration at the University of Kentucky.

His research interests include government

performance, managerial succession, and

citizen interactions with public services. He

has done extensive work on these topics in

two collaborative projects on English local

government and British central government.

E-mail: nicolai.petrovsky@uky.edu

Abstract : There are a number of influences on how long an agency head serves. The importance of particular

influences, in turn, depends on the prospective destination of the agency head: elsewhere in the public sector, the

private sector, or retirement. The authors estimate survival models of agency heads’ tenure using panel data on British

central government executive agencies from 1989 to 2012. Findings suggest that chief executives of poorly performing

agencies are encouraged to retire sooner. There is no evidence that a change in political control increases the risk of

any form of exit, suggesting that political pressure to leave is not substantial for this type of official. Outsiders—agency

heads recruited from outside central government—are relatively difficult to retain for a longer time, such that potential

shortfalls in suitable managers caused by retirements in an aging workforce may be difficult to make up by appointing

from this source.

Practitioner Points

• We analyzed the length of tenure of the heads of British central government executive agencies. This type of

organization, first established in 1989, involves a high degree of autonomy for agency heads.

• For our study, we built a database of these officials and the organizations they led, ranging from the first

agencies in 1989 until 2012.

• Heads of agencies that perform poorly relative to other agencies tend to retire somewhat sooner than others

and well before the usual retirement age. This is consistent with the policy behind the executive agency

organizational model as “performance-based organizations.”

• Regardless of actual performance consequences, stakeholders may demand symbolic managerial turnover. A

flexible personnel system like the British one for senior public managers is aligned with such demands.

• Agency heads recruited from outside central government are relatively difficult to retain for a longer time. If

there is a wave of retirements of senior career civil servants, it is a risky strategy to expect to replace them with

lateral recruitments from other levels of government and the private sector.

Nicolai Petrovsky

University of Kentucky

Oliver James

Alice Moseley

University of Exeter , United Kingdom

George A. Boyne

Cardiff University , United Kingdom

What Explains Agency Heads’ Length of Tenure?

Testing Managerial Background, Performance, and

Political Environment Effects

A major issue in public service management is

how long agency heads serve. Length of tenure

has important implications because if agency

heads serve only a short time, they may not be able

to contribute effectively to the organization they lead

(Hess 1998 ) or may be part of more general turnover

that damages performance (Bertelli and Lewis 2013 ).

In contrast, if they stay too long, their leadership

may stagnate and not keep pace with changes in the

organizational environment (Miller 1991 ). Theory

suggests that there are certain personal, organizational,

and environmental characteristics that predict the

time that senior managers will serve before leaving.

However, existing empirical research on public sector

executives has focused on organizational factors,

especially prior performance and political change

(Boyne et al. 2010a , 2010b ; Hood and Lodge 2006 ;

McCabe et al. 2007; Whitaker and DeHoog 1991 ).

We lack knowledge about the relevance of personal

characteristics such as previous work experience,

in contrast to the large amount of research on this

issue in the private sector (Finkelstein, Hambrick,

and Cannella 2009 ; Fredrickson, Hambrick, and

Baumrin 1998; Hambrick and Mason 1984 ; Karaevli

2007 ). The main exception has been recent work

looking at factors affecting the intention to exit of

public servants (Bertelli and Lewis 2013 ; Whitford

and Lee 2015 ). This work on turnover intention

has considerable relevance to understanding agency

heads’ tenure, but it is also important to look at actual

turnover.

Still, relatively little is known about realized turnover

at the managerial level. In their synthesis of research

on realized public employee turnover, Grissom, Viano,

and Selin ( 2016 ) identify three insights that are of

relevance to managerial turnover. First, it is difficult