NOT SO ‘ARM'S LENGTH’: REINTERPRETING AGENCIES IN UK CENTRAL GOVERNMENT
| DOI | http://doi.org/10.1111/padm.12089 |
| Date | 01 June 2014 |
| Published date | 01 June 2014 |
| Author | THOMAS ELSTON |
doi: 10.1111/padm.12089
NOT SO ‘ARM’S LENGTH’: REINTERPRETING
AGENCIES IN UK CENTRAL GOVERNMENT
THOMAS ELSTON
Administrative decentralization to government agencies (so-called ‘agencification’) has attracted
much attention in recent years, increasingly for its longevity or evolution after the ‘high’ manage-
rialism of the 1980s, and largely through a neo-positivist epistemology. Drawing on techniques of
narrative and discourse analysis, and a model of incremental ideational change, this article identifies
the necessity of supplementing those existing large-N analyses of agencification’s expansion and
decline with qualitative attention to the endurance of policy meaning. It demonstrates how the
original foundations of managerialism, civil service empowerment and decentralization from the
UK’s seminal ‘Next Steps’ agency programme are eschewed in contemporary reform discourse,
where agencification is instead advocated as centralized, politically proximate and departmental-
ized governance. This substantial reinterpretation of the arm’s-length concept not only challenges
existing claims of continuity in UK administrative policy, but also demonstrates the utility of
interpretive methods for exploring longevity in public management more widely.
INTRODUCTION
Decentralization from government ministries to arm’s-length public agencies – so-called
‘agencification’ – has attracted much attention in recent years (Pollitt et al. 2004; Chris-
tensen and Lægreid 2006; Smullen 2010; Verhoest et al. 2011; MacCarthaigh 2012; Elston
2013). In particular, this reflects the international ubiquity of ‘agency fever’ (Pollitt et al.
2001), and its collation and combination of multiple New Public Management (NPM) doc-
trines. Indeed, striving for bureaucratic disaggregation and de-regulation (Hood 1991),
organizational role purification (Christensen and Lægreid 2006), and a shift towards
ex-post, contract-based control (Hoggett 1996), the agency solution epitomizes several key
tenets of new managerialism. Pursued in some measure across countries as diverse as the
Netherlands, New Zealand, Japan, and the UK, agencification is amongst NPM’s ‘most
frequently adopted and far-reaching’ policies (Moynihan 2006, p. 1029).
Given this pervasiveness and intimate paradigmatic association, recent analysis of
managerialism’s post-2000 endurance has necessarily deliberated the longevity of agenci-
fication. Claiming NPM’s outright substitution by ‘digital-era governance’, for example,
Dunleavy et al. (2006) cite diminished enthusiasm amongst transnational reform advocates
such as the OECD, as well as the second-generation, anti-fragmentation measures initiated
by governments seeking better control and ‘join-up’. More cautiously, Christensen and
Lægreid (2007, p. 11) identify a layering of new ideals over enduring managerialist pre-
cepts, arguing not for a paradigm shift, but for ‘a change of emphasis away from structural
devolution, disaggregation and single-purpose organizations’. More delimited national
and comparative analyses of agencification have broadly supported this latter claim of
‘post-NPM’ evolution rather than revolution, reporting continuing administrative disper-
sal, tempered mainly by agency mergers, new coordination and oversight mechanisms,
and only limited formal de-agencification (see Verhoest et al. 2011; MacCarthaigh 2012;
Elston 2013).
Thomas Elston is at the Blavatnik School of Government, University of Oxford, UK.
Public Administration Vol. 92, No. 2, 2014 (458–476)
©2014 John Wiley & Sons Ltd.
REINTERPRETING AGENCIES IN UK CENTRAL GOVERNMENT 459
Focused on the preservation of devolved organizational landscapes, much of this
longitudinal research involves large-N ‘population’ analysis. Indeed, of the three emergent
modes of agency theorizing identified by Pollitt et al. (2004, pp. 12–18) – ‘traditional’
organization science, micro-economic, and interpretive – the former has come to dominate,
largely by virtue of burgeoning population studies (see examples in Verhoest et al.
2011; MacCarthaigh and Roness 2012). Borrowing assumptions and techniques from
industrial organization ecology (Hannan and Freeman 1989), population analysis attends
to aggregate administrative systems.
‘Agency’ is reified as an objective classification which engenders some common struc-
tural influence across otherwise disparate governing processes and outcomes. While thus
perpetuating the undifferentiated, catch-all logic by which NPM’s ‘best-practice’ reforms
were often pursued during the 1980s and 1990s, the meta-theoretical assumption of a
single, essential, and ‘univocal’ agency function contrasts markedly with the founding
premise of Pollitt et al.’s (2004) third, interpretive branch of theorizing, for which social
and political phenomena are modelled as intangible structures of meaning, lacking in
fixity and sustained through perpetual social re-accomplishment in confined ideational
contexts. This alternative epistemology indicates how ‘public management ideas that
carry the same identifying label can mask variation in the understanding of the policy’
(Moynihan 2006, p. 1029). In short, the meaning of ‘agency’ is locally and temporally
contingent, susceptible to both synchronic and diachronic reinterpretation.
Despite growing interest in post-empiricist policy and organization studies (for
example, Burrell and Morgan 1979; Hawkesworth 1988; Wagenaar 2011), interpretive
agencification research remains incipient. Exceptionally, in exploring the interplay of
local culture and transnational reform discourse, Moynihan (2006) and Smullen (2010)
identify distinct cross-national ‘translations’ of the agency idea, while Smullen (2004) also
points to subnational reframing within a single polity. Principally, therefore, this still
nascent research literature explores synchronic reinterpretations of the agency solution
across different locales, rather than diachronic instability over time. As such, interpretive
methods have yet to contribute substantively to the question of longevity in agency policy,
practice, and paradigm.
Against this backdrop, the present article demonstrates the necessity of supplementing
longitudinal, large-N population analysis with attention to the evolution of policy mean-
ings. Drawing on the UK’s seminal ‘Next Steps’ case of agencification, it traces how the
implicit understandings behind an ostensibly enduring organizational classification have
reshaped over time. Not only does this indicate the limited sustenance of a mainstay of
the UK’s NPM project, but it also highlights how inductive attention to, and comparison
of, tacit and embedded knowledge makes for fuller analysis of continuity and change in
public management more widely.
Initially, the article builds a conceptual and methodological framework, drawing on
constructivist meta-theory, a model of political ideas as composite discursive structures,
and insights from narrative and discourse analysis. Thereafter, textual sources pertaining
to the UK’s 1988 Next Steps programme and 2010 Public Bodies Reforms are examined for
their separate evocations of the agency idea. Overall, the original emphasis on managerial
empowerment, devolution and de-politicization is found to be largely absent from the
contemporary reconstruction, being replaced by counter-themes of ministerial control
and centralization. Together, this indicates the latter-day departmentalization of the
once arm’s-length agency concept – a modification of tacit meaning which has occurred
without legal redefinition or overt policy change. Thus overlooked by large-N system
Public Administration Vol. 92, No. 2, 2014 (458–476)
©2014 John Wiley & Sons Ltd.
To continue reading
Request your trial