Description
Changes in public sector accow~tingin a number of OECD countries over the 198Bs were central to the
rise of the “New Public Mulaganent” (NPM) and its asso&md doctrks of public accoumabihty and
organizitiooal best practice. lhis paper d&uases theriseofNPMuulalvmativetothetnditionofpuMic
accountabilityembodiedin pmgmssive-era public admin&ra&n ideas. It argues that, in spite of allegations
ofintem?tionalintionurdtheadoptionofanewgiobalp;udigminpuMic mylaganent,d==w=
considerable variation in the extent to which different OECD countries xlopted
Ac an#i n& ol gaf uul l ons and Soc i et y, Vol . 20, No. U3, pp. 93-109, 1995
Ekvicr Science Ltd
Printed in Great Britain
0361~36Bzl95 19.50+0.00
THE “NEW PUBLIC MANAGEMENT” IN THE 1980s:
VARIATIONS ON A THEME’
CHRISTOPHER HOOD
London School of Economics and Political Science
Changes in public sector accow~ting in a number of OECD countries over the 198Bs were central to the
rise of the “New Public Mulaganent” (NPM) and its asso&md doctrks of public accoumabihty and
organizitiooal best practice. lhis paper d&uases theriseofNPMuulalvmativetothetnditionofpuMic
accountabilityembodiedin pmgmssive-era public admin&ra&n ideas. It argues that, in spite of allegations
ofintem?tionalintionurdtheadoptionofanewgiobalp;udigminpuMic
mylaganent,d==w=
considerable variation in the extent to which different OECD countries xlopted NPM over the 19t3Os. It
further argues that conventiortal expMm&u of the rise of NPM (“w, party pohticai incumbency,
etonomic~tecorduldgovanment~)seanhudtosust?inevcn6romarrl?tivclybrief
inspectionofsuchaossnationaldua~ucavailable,uldthatulaplul?tionbvedoninitialendowmcnt
maygiveusaditkrentpeqectiveonthosechanges.
Over the 1980s there was a move in a number
of OECD countries towards the New Public
Management (NPM). Central to this change in
modes of public management was a shift towards
“accountingization” (a term coined by Power 81
Laughlin, 1992, p. 133).’ This development can
be claimed to be part of a broader shift in
received doctrines of public accountability and
public administration At the same time, account-
ing changes formed an important part of the
assault on the progressive-era models of public
accountability (cf Hal&an & Wettenhall, 1990).
For progressive public administration,2 demo-
cratic accountability depends on limiting corrup-
tion and the waste and incompetence that are
held to go with it (cf. Karl, 1963, p. 18). The
assumption is that politicians are inherently
venal, using their public office wherever pos-
sible to enrich themselves, their friends and
relations, and that reliance on private-sector
contracting for public services inevitably leads
to high-cost lowquality products, either because
of corrupt influence on the contract-awarding
process or because the public contract market
will come to be controlled by organized crime,
or both. Whether these assumptions can be
safely dispensed with in the wealthy OECD
countries of today is a matter for debate.
From those assumptions, the accountability
paradigm of progressive public administration
(hereafter PPA for convenience) put heavy
stress on two basic management doctrines. One
of those doctrines was to keep the public sector
sharply distinct from the private sector in terms
l This paper was presented in an earlier form to the Workshop on “Changing Notions of Accountability in the UK Public
Sector”, SE, December 1991, and to the ElASM Workshop on Accounting, Accountability and the “New European Public
Sector”, Helsinki, September 1992.1 am grateful for comments and criticisms received on those occasions, and particularly
to Peter Miller for helpful suggestions.
’ Broadly, “accountingization” means the introduction of ever-more explicit cost categorization into areas where costs
were previously aggregated, pooled or undefined.
* That is, the style of public administration that emerged in the “progressive era” of the late nineteenth and early twentieth
centuries.
93
94
C. HOOD
of continuity, ethos, methods of doing business,
organizational design, people, rewards and
career structure. The aim, in Beatrice Webb’s
words (Barker, 1984, p. 34) was for a “Jesuitical
corps” of ascetic zealots. The other doctrine was
to maintain buffers against political and mana-
gerial discretion by means of an elaborate
structure of procedural rules designed to
prevent favouritism and corruption and to keep
arms-length relations between politicians and
the entrenched custodians of particular public
service “trusts”.3
This organizational model attracts more deri-
sion than analysis today (cf. Osborne & Gaebler,
1992). In fact, it reflects an underlying metaphor
of trustee and beneficiary (John Locke’s meta-
phor for government) and involves a complex
mix of high-trust and low-trust relationships, with
the accompanying accounting rules reflecting
degrees of trust. Within the “Jesuitical corps” of
the public service were many high-trust rela-
tionships (for example, in conventions of mutual
consultation or action on the basis of word-of-
mouth agreements across departments), the
costs of which were not accountingized. The
implicit assumption is that such high-trust,
non-costed behaviour lowers transaction costs
within the public sector and makes it more
efficient than it would be if each action had to
be negotiated and costed on a low-trust basis.
However, PPA also embodied many low-trust
relationships, particularly in areas where the
Jesuitical corps faced the corrupting forces of
the world outside, notably the award of con-
tracts, recruitment and staffing, as well as the
handling of cash, where distrust prevailed and
elaborate records had to be kept and audited.
In the place of the PPA model came New
Public Management or NPM (cf. Aucoin, 1990;
Hood, 1987, 1990a, b, 1991; Dunsire & Hood,
1989; Hood & Jackson, 1991; Pollitt, 1993;
Pusey, 199 1). NPM involved a different concep-
tion of public accountability, with different
patterns of trust and distrust and hence a
different style of accountingization. The basis of
NPM lay in reversing the two cardinal doctrines
of PPA; that is, lessening or removing differences
between the public and the private sector and
shifting the emphasis from process account-
ability towards a greater element of account-
ability in terms of results. Accounting was to be
a key element in this new conception of
accountability, since it reflected high trust in
the market and private business methods (no
longer to be equated with organized crime) and
low trust in public servants and professionals
(now seen as budget-maximizing bureaucrats
rather than Jesuitical ascetics), whose activities
therefore needed to be more closely costed and
evaluated by accounting techniques. The ideas
of NPM were couched in the language of
economic rationalism, and promoted by a new
generation of “econocrats” and “accountocrats”
in high public office.
The term NPM was coined because some
generic label seemed to be needed for a general,
though certainly not universal, shift in public
management styles. The term was intended to
cut across the particular language of individual
projects or countries (such as the French ‘Projet
de Service”, the British “Next Steps”, the
Canadian “Public Service 2000”). The analogy
is with terms like new politics, new right, and
new industrial state, which were invented for a
similar reason.4
As with the disappearance of the dinosaurs,
there is no single accepted explanation of this
alleged paradigm shift. In fact, emerging expla-
nations roughly parallel the major contending
theories of the dinosaurs’ extinction. Some
.3 An organizational structure which could clearly be classed as “hierarchist” in the cultural theory of Mary Douglas ( 1982)
and her followers (Thompson et al.. 1990).
4 The term “new” does not imply that NPM doctrines appeared for the first time in the 1980s (any more than the “New
Learning” of the fifteenth and sixteenth centuries involved the first discovery of Latin and Greek). Many NPM doctrines
repackage ideas which have been in public administration since its earliest beginnings. Nor must NPM be confused with
the “New Public Administration” movement in the U.S.A. in the late 1960s and early 1970s. which achieved no real
mainstream influence (see Marini, 1971).
THE NEW PUBLIC MANAGEMENT 95
accounts stress “habitat loss” for the old style
arising from post-industrial technology, pro-
ducing a wholly new model of public admini-
stration built around electronic data handling
and networking, providing many new niches for
accountingization and lowering its direct costs
(see Taylor & Williams, 1991, p. 172; Taylor,
1992). Some see the demise of the old model
as the result of a sudden shock, with New Bight
ideas about organizational design coming as a
meteorite from out of the blue (as in Quirks,
1988, idea-centric account of 1980s’ deregula-
tion). Some see PPA’s fate as a self-induced
extinction, as older control frameworks and
accounting practices came to degrade the values
which they were designed to promote (see
Painter, 1990, p. 77; Hirschman, 1982). And yet
others interpret the change as caused by a new
set of predator interests, such as accounting
firms and management consultants, hunting PPA
into extinction (see Dunleavy, 1985, 1986,
1991).
contains important variations, and that no
account of the shift from the progressive public
administration model to NPM can be satisfactory
unless it can account for international leaders
and laggards.
NEW PUBLIC MANAGEMENT; THEMES
AND VARIATIONS
Themes
However, before such “extinction science”
can be developed, we need to be satisfied that
some general extinction has actually taken
place, and that the new life-form of NPM is
everywhere supplanting PPA. Such claims are
indeed commonly made both by practitioners
and by academic commentators. Aucoin (1990,
p. 134) for instance, asserts that: “What has been
taking place in almost every government in
developed political systems and highly institu-
tionalized administrative states is a new emphasis
on the organizational designs for public manage-
ment . . This internationalization of public
management parallels the internationalization of
public and private sector economies.” Similarly,
Osborne & Gaebler (1992, pp. 322-330) write
of NPM as a new “global paradigm”, claiming
that transition to the new paradigm is inevitable
“just as the transition from machine rule to
Progressive government was inevitable” (p.
325).
The doctrines of public sector management
encompassed by NPM have been variously
described by different commentators (such as
Aucoin, 1990; Hood, 1991; Pollitt, 1993) and
some have identiBed diRerent phases in the
development of NPM. However, there is still a
good deal of overlap among the different
accounts of what NPM entailed. For example,
the idea of a shift in emphasis from policy making
to management skills, from a stress on process
to a stress on output, from orderly hierarchies
to an intendedly more competitive basis for
providing public services, from fared to variable
pay and from a uniform and inclusive public
service to a variant structure with more
emphasis on contract provision, are themes
which appear in most accounts.
Most commentators have associated NPM
with approximately seven dimensions of change,
which are summarized in Table 1, together with
their associated doctrines, and some speculative
ideas about their implications for accountingiza-
tion. The elements relate to the two cardinal
elements of PPA already noted, in that the first
four elements of Table 1 relate to the issue of
how far the public sector should be distinct from
the private sector in its organization and
methods of accountability, and the last three
broadly relate to the issue of how far managerial
and professional discretion should be fenced in
by explicit standards and rules. The seven
elements are as follows.
If we accept such claims, we would expect ( 1) A shift towards greater disuggmgution of
to see a process of international convergence public organizations into separately managed
and diffusion of NPM ideas in public administra- “corporatized” units for each public sector
tion. But this paper argues that the inter- “product” (each identified as a separate cost
nationalization of the NPM model at least centre, with its own organizational identity in
% C. HOOD
TABLE 1. Doctrinal components of new public vt
No. Doctrine T)piCd Replaces OpentiOMl Some possible
justification significance accounting
imDliatiOnS
PS dtsftncttwness
Unbundling of the
Ps into corporatized
units organized by
product
More contract-
based competitive
provision, with
internal markets and
term contracts
Stress on private-
sector styles of
management
practice
More stress on
discipline and
liugality in resource
USe
Rules us dtscwtton
5. More emphasis on
visible hands-on top
management
6. Explicit formal Accountability
measurable stan- means clearly stated
dards and measures aims; efficiency
of pefonnce and needs hard look at
success goals
7. Greater emphasis
on output controls
Need for greater
stress on results
Make units
manageable, and
focus blame; split
provision and
production to create
anti-waste lobby
Rivahy as the key to
lower costs and
better standards;
contracts as the key
to explicating
performance
SGMdd.3
Need to apply
proven private-
sector management
tools in the public
sector
Need to cut direct
costs, raise labour
discipline, do more
with less
Accountability
requires clear
assignment of
responsiblity not
difision of power
Belief in uniform Erosion of single
and inclusive PS to service
avoid underlaps and employment; atms-
overlaps in length dealings;
accountability devolved budgets
Unspecified
employment
contracts, open-
ended provision,
linking of purchase,
provision,
production, to cut
transaction cost
Stress on PS ethic
fixed pay and hiring
rules, model
employer
orientation
centralized
personnel structure,
jobs for life
Stable base budget
and establishment
norms, minimum
standards, union
vetoes
Paramount stress
on policy skiUs and
rules, not active
management
Qualitative and
implicit standards
and norms
Stress on procedure Resources and pay
and control by based on
collibration performance
activity
Source: adapted from Hood (1991, pp. 4-5).
More cost centre units
Distinction of
primvy and
secondary public
service labour force
Move from double
imbalance PS pay,
career service,
unmonetized
rewards “due
process” employee
entitlements
Less Primary
employmen& less
job security, less
producer-friendly
style
More stress on
identifying costs and
understanding cost
structures; so cost
data become
commercially
confidential and
cooperative behaviour
becomes costly
Private-sector
accounting norms
More stress on the
bottom line
More freedom to
manage by
disaetionaty power
Erosion of self-
management by
professionals
Fewer general
procedural constraints
on handling of
contracts, cash, staff;
coupled with more use
of fmancial data for
management
accountability
Pertixmance
indicators and audit
Move away from
detailed accounting
for particular activities
towards broader cost
centre accounting;
may involve blurring
oftundsbrpayandfor
THE NEW PUBLIC MANAGEMENT 97
fact if not in law, and greater delegation of
resource decisions, in a movement towards
“one-line” budgets, mission statements, business
plans and managerial autonomy). The corpora-
tized style contrasts with the PPA style of
providing all public services through “semi-
anonymized” organizations within a single
aggregated unit, with detailed service-wide
rules, common service provision in key areas of
operation, detailed central control of pay
bargaining and stafhng levels.
(2) A shift towards greater competition both
between public sector organizations and bet-
ween public sector organizations and the private
sector. The aim for a more competitive style
contrasts with the PPA style of ascribing semi-
permanent “ascribed” roles to public sector
organizations; that is, captive markets which are
indefinitely assigned to particular “prestige”
producers.
(3) A move towards greater use within the
public sector of management practices which
are broadly drawn from the private corporate
sector, rather than PPA-style public-sector-
specific methods of doing business. Examples of
the latter include “model employer” aspirations
to set an example to, rather than to follow the
lead of, private-sector employers in matters of
pay and conditions of employment (for example,
in equal opportunity or employment of disabled
persons) and the traditional “double imbalance”
pay structure of public administration, in which
lower-level statf tend to be relatively highly paid
compared to their private-sector counterparts
and top-level staff are relatively low-paid (cf.
Sjolund, 1989).
(4) A move towards greater stress on disci-
pline and parsimony in resource use and on
active search for finding alternative, less costly
ways to deliver public services, instead of laying
the emphasis on institutional continuity, the
maintenance of public services which are stable
in “volume terms” and on policy development.
( 5) A move towards more “‘hunds-on manage-
ment” (that is, more active control of public
organizations by visible top managers wielding
discretionary power) as against the traditional
PPA style of “handsofF management in the
public sector, involving relatively anonymous
bureaucrats at the top of public-sector organiza-
tions, carefully fenced in by personnel manage-
ment rules designed to prevent favouritism and
harassment.
(6) A move towards more explicit and
measurable (or at least checkable) stanahrds
of performance for public sector organizations,
in terms of the range, level and content of
services to be provided, as against trust in
professional standards and expertise across the
public sector. The old PPA style involved low
trust in politicians and managers but relatively
high trust in professional expertise, both in a
“vertical” sense (that is, up and down the
organizational ladder, or between “principals”
and “agents” in the new legal+conomic lan-
guage of the economic rationalists) and in a
“lateral” sense (that is, across difyerent units of
the public sector; cf. Fox, 1974, pp. 72-84,
102-l 19).
(7) Attempts to control public organizations
in a more “homeostatic” style according to pre-
set output measures (particularly in pay based
on job performance rather than rank or educa-
tional attainment), rather than by the traditional
style of “orders of the day” coming on an ad
hoc basis from the top, or by the subtle balancing
of incompatible desiderata in the “collibration”
style of control identified by Dunsire (1978,
1990) as central to orthodox bureaucratic
functioning.
These doctrines of NPM link to recurrent
debates about how public administration should
be conducted, which stretch back at least as far
as the major disputes between “legalists” and
“Confucians” in the Chinese mandarinate over
2000 years ago (see Kamenka, 1989, pp.
38-39). How far the public sector should be
insulated and clearly separated from the private
sector in matters of handling business and staff,
and how far business should be conducted by
professional discretion rather than by pre-set
rules or standards, are issues which go to the
heart of most doctrinal disputes in public admini-
stration, including such major waves of classic
public administration thought as the ideas of the
German cameralists from the mid-sixteenth
98
C. HOOD
century (Small, 1909) the nineteenth-century
British utilitarians (Hume, 1981) and the turn-
of-the-century American progressives (Ostrom,
1974). Such doctrines also have profound
implications for how public sector accounting
is conceived, in the sense of what records are
kept, how they are used, and what is costed and
measured.
This summary list of course oversimplifies,
and there are many interesting counter-trends.
Examples of such counter-trends include: the
unfashionability of the traditional public enter-
prise model in conventional market sectors of
the economy, coupled with the vigorous adop-
tion of that model for non-marketed public
services in several countries; the weakening of
older doctrines of metaphytic competition (i.e.
public versus private providers; Corbett, 1965)
as against doctrines of market testing by
franchising; the weakening of trust in profes-
sionals while strengthening the hand of mana-
gers. Certainly, there is no logical necessity for
a public management system to change in all of
these seven respects at once. Many variations
are possible.
Variations
There are no systematic cross-national studies
showing degrees of variation in public manage-
ment reform in a robust and reliable way. The
literature in the area is long in anecdote and
general commentary but short on systematic
comparison, and comes close to being a data-
free environment. There are only isolated
fragments and relatively low-grade comparative
data from sources like OECD public manage-
ment reports and cross-national consultancy
reports such as the Price WaterhouseKranfield
study of comparative pay flexibility (Hegewisch,
1991).
But even such fragmentary sources are suffi-
cient to show that not all OECD countries moved
to adopt NPM principles to the same extent during
the 1980% and that there were marked diffe-
rences even within similar family groups such
as the English-speaking “Westminster-model”
countries (cf. Hood, 1990~). It is particularly
notable that some of the OECD’s showcase
economies, Japan, Germany and Switzerland,
seem to have put much less emphasis on
adopting NPM-type reforms (on the seven
dimensions indicated in Table 1) in the 1980s
than countries like Sweden, New Zealand or the
U.K. But it would be hard to argue that those
countries were closest to the NPM model at the
outset, particularly in respect of use of private
sector-style management practices, hands-on
management or output controls.
For example, the NPM tendency to decentra-
lize personnel management (such as hiring and
job classifications) to operating units away from
central oversight agencies was not a marked
tendency in Japan, where the National Personnel
Authority was if anything strengthened rather
than weakened over the 1980s. Administrative
reform received much attention, in the form of
the three reform commissions over that decade
(modelled on the famous 1937 U.S. “Brownlow”
Committee on Administrative Management),
but the accent seems to have been more on
privatization, deregulation and tax reform than
on the principles of NPM, with the exception
of a small measure of corporatization in the form
of more freedom for Ministries to reorganize
themselves without specific authority from the
Diet. And whereas doctrines of “pay for perfor-
mance” took a strong hold in countries such as
Sweden, Denmark, New Zealand and the U.K.,
there was no equivalent movement in Germany
(partly because pay for performance poten-
tially conflicts with the Basic Law doctrine of
equality of pay across particular grades in the
public service). Indeed, neither Germany nor
Switzerland made any major changes in their
public administration at federal level in the
1980s; rather, Vewaltungspjlege (a quiet period
of cultivation after the reforms of the 1960s which
embraced policy evaluation) was a common
watchword in Germany over that decade.
But even within the group of countries in
which more emphasis seemed to be placed on
public management reforms, it is not clear that
the direction of change was the same. For
instance, the “free commune” experiment in
Norway in 1987 and the French move to far-
reaching territorial decentralization in 1983
THE NEW PUBLIC MANAGEMENT 99
contrasts sharply with centralizing tendencies
in the U.K. over the 1980s. Late 1980s reforms
in the U.K. and New Zealand were aimed at
separating policy setting and service provi-
sion (on the grounds that Ministers were not
equipped to be managers), while the Australian
Commonwealth government took measures
intended to strengthen the capacity of Ministers
to manage.
Any scoring of variations in the adoption of
NPM in the seven dimensions discussed above
must necessarily be highly impressionistic.
Ideally, we would need both a reliable method
of locating a country’s initial state at the start
of the period in question and the extent of
movement over the period (given that a
“backward” case might show dramatic change,
and yet still be behind an apparently “static”
country which started from a higher initial
emphasis on the NPM style). At present, we do
not have a cross-national information source
which could reliably show either sort of
variation.
From the fragmentary literature on public
management reform over the 1980s the high
NPM group in the OECD countries would be
likely to include Sweden, Canada, New Zealand,
Australia and the U.K. with France, Denmark,
the Netherlands, Norway and Ireland also
showing a number of marked shifts in the
direction of NPM. At the other end, the low NPM
group would be likely to include Germany,
Greece, Spain, Switzerland, Japan and Turkey.
These impressions are consistent with the
country reports submitted to the OECD’s survey
of public management developments in 1988
and 1990, supplemented in 1991 (OECD 1988,
1990, 1991). Obviously, these reports are
seriously contaminated in a number of familiar
ways, mainly because they reflect what the
correspondents used by OECD in each country
thought it relevant or politic to record (although
there was, of course, a general check-list issued
by PUMA for these exercises), rather than what
a single overall observer might have noted.
Moreover (until such time as the OECD’s public
management profiles are greatly developed), it
indicates what was on the government agenda
for change rather than the “initial endowment”
of each system.
Even so, OECD public management reports
can give us a rough indication of the officially
perceived agenda of public management change
in the 198Os, and serve as a starting point for
discussion of variations. The country reports for
1988 and 1990 were carefully examined, and a
rough score awarded to each country under
each of the seven points of public management
doctrine which were mentioned above as
components of new public management. A score
of 2 was given for developments reported as
being in place on each of the dimensions, 1 for
developments under active discussion or experi-
mentation and 0 for nothing reported in the
area.
Clearly, such an exercise is only useful for
identifying outliers, and not for making any fine
difTerentiations. The extremes were taken as
cases whose overall scores (summed across all
seven dimensions) were more than one standard
deviation away from the mean. The high and
low cases identified in this way fitted sufficiently
well with other impressionistic views of varia-
tion in NPM over the 1980s to serve as a basis
for a discussion as to what might be responsible
for putting a country at one or other of the
extremes.
WHY THE VARIATIONS?
As noted earlier, there are several different
possible ways of explaining the rise of NPM. This
section explores four conventional accounts of
public sector change (relating such change to
“Englishness”, party politics, government size
and macroeconomic performance) before sug-
gesting a further interpretation couched in
terms of institutional endowments.
‘English awfulness”?
Some commentators (notably Pollitt, 1993)
have implied that NPM was mainly an Angle+
American phenomenon of the ReaganThatcher
era. But this view seems difficult to sustain. If
nothing else. it ignores the high degree of
100
C. HOOD
emphasis placed on NPM in South Africa, Hong
Kong, Australia and New Zealand (for the latter,
see Pusey, 1991; Scott et al., 1990; Yeatman,
1987).
However, it might be possible to broaden tire
view of NPM as an Anglo-American preoccupa-
tion to the idea that it reflects what Castles
ironically calls “the awfulness of the English”.
By this phrase, Castles means the relatively
poor economic performance and arrested deve-
lopment of welfare state policies which, he
claims, characterize the English-speaking coun-
tries (Castles, 1989; Castles & Merrill, 1989, pp.
181-185). It is, for example, noticeable that the
high scorers on NPM emphasis are mostly English-
speaking countries (and hence clearly candidates
for “English awfulness” ti Zu Castles). The low
scorers, in contrast, are all non-English-speaking
countries.
Moreover, an “English awfulness” explanation
might fit with a view of NPM as representing
international convergence on a common public
management style. On Castles’ analysis, the
English-speaking countries lost their formerly
distinctive “high-direct employment” feature of
public management between the 1960s and the
198Os, and NPM could be interpreted as part of
that process of coming into line with non-
English-speaking countries. Particularly with
respect to the practice of accounting tech-
niques and management consultancy, diflu-
sion of the new NPM ideas might have been
expected to spread more readily across coun-
tries with the same language and similar legal
traditions.
But it seems too simple to attribute NPM to
English awfulness alone. For example, Sweden
appears as a high scorer on NPM emphasis, which
would be particularly damaging for an interpre-
tation built on a Castles-type English awfulness
factor. Moreover, Denmark, the Netherlands and
France are also cases which score relatively high
on NPM emphasis, and two of them (Denmark
and the Netherlands) reported strong develop-
TABLE 2. NPM emphasis and political incumbency
Political incumbency emphasis
NPM
Emphasis “LeW “Centre”
” WW
Hi Sweden
Medium
Lo
France
Greece
NZ
Austria
Denmark
Finland
ItafY
Netherlands
Portugal
USA5
Germany
JaPm
Snain Switzerland Turkev
Sources: analysis bved on OECD PUMA reports and on
political incumbency data for OECD countries drawn from
Gorvin (1989) and Keestng’s Contem~ Ar c hf ves.
ment of variable pay in recent years to the 1990
Price WaterhouseICranfield project (Hegewisch,
199 1, Table 3). NPM seems to be more than just
another “English disease”.
Party politics
Some commentators explain PPA’s demise in
rather similar terms to the predator theorists of
the dinosaurs’ extinction. The notion is that the
old structure has been subverted by the deve-
lopment of “New Right” interests who stand to
benefit in various ways from dismantling the
PPA model.
At one level, the demise of traditional PPA is
often attributed to the advent of New Right
government in the 1980s and particularly to
the infiuence of Ronald Reagan and Margaret
Thatcher, who aimed to roll back big govern-
ment and state-led egalitarianism and welfarism,
and to remould what was left of the public sector
in the image of private business (see Pollitt,
1993). lf NPM was sparked by such figures, we
would expect its development to be most
marked in countries which were governed by
right-wing parties during the 1980s.
’ It is obviously debatable whether U.S.A. should be scored as “centre” or “right” for these purposes. In many “does party
matter” studies. it is excluded, on the grounds that it is essentially unclassifiable. Given its separation of powers, it is here
counted asacentre-leftcoalitionfor 1980,acentre-right coalition for 1981 to 1986andacentristcoalitionfor 1987 to 1990.
THE NEW PUBLIC MANAGEMENT 101
Table 2 gives a rough indication of the extent
to which OECD countries were governed during
the 1980s mainly by parties on their left, their
right or their centre. For this exercise, a score
of i- 1 was given to each OECD country for each
year of incumbency in government by a political
party to the right of that country’s political
spectrum, and a score of -1 for each year of
incumbency by a party to the left of that
spectrum; 0.5 was given for each year of
incumbency by a centre-right coalition and
-0.5 for each year of incumbency by a centre-
left coalition, with 0 for a grand coalition (as in
the Austrian case). Countries which score hi
(more than 1 S.D. above the mean) are taken as
“right” in incumbency terms, and countries
which score lo (more than 1 S.D. below the
mean) are taken as “left” in incumbency terms.
“Centre” scorers are the rest.
NPM scorings are based on the country
reports to OECD on Public Management (1988
and 1990), with a maximum, score of 2 on each
of seven dimensions of doctrine, as described
earlier. “Hi” scorers are those whose overall
score is more than 1 standard deviation (S.D.)
above the mean score; “lo” scorers are those
whose overall score is more than 1 S.D. below
the mean score. “Medium” scorers are the
rest.
This exercise is useful only for first approxi-
mations, and it may be that a better test would
be a series of “before” and “after” looks at cases
of changes in government. Scoring political
incumbency involves problems which are fami-
liar in the “does politics matter?” debate over
macroeconomic policy in political science.
There is no established method for comparing
degrees of “rightness” and “leftness” ac?wss
countries, and there are clearly dilYerent “qual i -
ti es” of rightness and leftness (for example,
participatory versus hierarchical emphases in
socialism). Presidential and federal systems
clearly cause complications (for example, the
I7.S. would count as “right” only in Presidential
terms for the 1980s). Coalition/PR system cases
tend to bunch in the middle, so that the outliers
rend to be non-PR systems like France and Japan,
which may get disproportionate weight in the
analysis. All that such a scoring exercise does is
to give us a first cut as to whether OECD
countries were governed during the 1980s
mainly by parties on their left, their right or
their centre.
Crude as the data are, however, Table 2 shows
up the difiiculties in the popular idea that NPM
was closely associated with incumbency by
“right” governing parties in the 1980s. Sweden
is the most obvious misfit for such an idea. It is
a country which shows apparently high NPM
emphasis during the 198Os, but also scores fairly
high for left political incumbency with eight
years out of the decade under Social Democratic
governments. Indeed, Sweden is conventionally
taken as the leading case of the social demo-
cratic alternative to liberal capitalism. And at
the other extreme, there are unambiguously
“right” cases, like Japan and Turkey, which seem
to score distinctly low on the NPM emphasis
scale.
Of course, such results are only surprising if
we expect the incumbency of dserent political
parties to lead to different public policy
outcomes. Other analyses of party competition
might lead to different expectations. An example
is Scharpfs ( 1987) “nested game” model of the
dynamics of party competition, in which two
rival “left” and “right” parties compete for the
fickle favours of floating middle-ground voters
against a set of variant macroeconomic condi-
tions produced by a game between incumbent
governments, labour unions and central banks.
Modifying that framework only slightly, we
could posit a “party competition” game played
between right and left wing parties (ri la
Scharpf) and a “public management game”
played between incumbent governments and
public managers/top bureaucrats or profes-
sionals (substituted for Scharpfs macroeconomic
policy game played between governments and
labour unions). In the second game, incumbent
politicians choose between tough and tender
approaches to public managers, and the latter
choose between a cooperative and uncoopera-
tive approach to politicians. Table 3 outlines
such a game. Clearly, politicians would prefer
to be in cell ( l), bureaucrats and managers in
102 C. HOOD
TABLE 3. Politician-bureaucrat public management game
Bureaucrats/
managers/
Politicians
public servants Tender Tough
Cooperative
(1) (2)
Outcome: smooth Outcome:
running medium cost politicians
public management exploit public
service ethos for
cheap and
effective public
management
Noncooperative
(3) (4)
Outcome: Outcome: high cost
bureaucrats exploit administrative
politicians’ chaos
goodwill for high
cost public
management
cell (3); such a game therefore has a tragic bias
away from cell ( 1) and towards cell (4).
If such a model of party competition is
adopted, it becomes quite conceivable that
incumbent social-democratic governments (such
as those in Sweden or New Zealand in the 1980)
might go just as far as “bourgeois” governments
(and possibly even further) in moving towards a
tough approach to public management, because
they have nowhere else to go, and need to work
harder to establish credibility in this area with
wavering middle-class voters (adapting Scharpfs
ideas of asymmetrical choice by the middle
socioeconomic group).
From such evidence, it would seem that if the
rise of the New Right influenced the demise of
the old PPA model, it was not simply through
right-wing parties holding office. It looks more
like a general process of policy diffusion than a
process in which policies vary with party
incumbency. A modified version of the rise of
the New Right explanation would therefore
need to place more stress on the eclipsing of
the old public administration coalition (parti-
cularly the public service labour unions) by an
NPM coalition drawn from accounting firms,
financial intermediaries, management consul-
tants and business schools, along the lines of
Patrick Dunleavy’s (1986) account of how
policy booms develop through a coalition of
professional and corporate interests.
Alternatively (perhaps additionally), the change
might be attributed to the recolonization of the
public service from inside rather than to an
assault from outside. Some contemporary com-
mentators on the rise of NPM, notably Yeatman
( 1987, pp. 350-351) and Pusey ( 1991) have
applied a form of “new class” analysis to the
process, in the tradition of Bumham’s (1942)
analysis of managers as a new class. They claim
that the upper echelons of the public service
are increasingly being occupied by a new class
of “econocrats” (and perhaps “accountocrats”,
certainly in the New Zealand case), who were
not fired up into public service ideals by close
experience of the rigours of ordinary life in the
depressed 1930s or Word War II and whose
education in narrow neoclassical economics
(uncontaminated by the humanities or even by
the other social sciences) in elite universities is
claimed to make them natural sympathizers with
New Right ideas.
It may be too that such a new generation of
econocrats and accountocrats would be much
more prone to “bureau-shaping”, as analysed by
Patrick Dunleavy ( 1985, 199 1); that is, the
propensity of top public managers to aim for
high-status analytic work in collegial elite units
and to distance themselves from front-line
supervisory roles in favour of a “super-control”
position which offers more job satisfaction and
less tedious routine. Once bureaucrats adopt
such preferences, there is nothing against their
interests in enthusiastically cutting service
delivery budgets down the line, or in breaking
up and deprivileging the world of public service
delivery, so long as the power and status of
central agencies is retained or augmented.
A response to fi scal stress and poor
macroeconomi c pwformunce?
Another common way of interpreting the
demise of the old PPA model is to link it with
changing social conditions, akin to the “loss of
habitat” theory of dinosaur extinction. NPM is
often interpreted as a response to fiscal stress
and resistance to extra taxes. Underlying the measures of government size can conceal as
onset of such fiscal stress may be the changes much as they reveal (Peters & Heisler, 1983).
in income level and distribution, weakening the But if we look at four conventional measures of
“Tocqueville coalition” for government growth government size (that is, government employ-
in the electorate; that is, an electoral majority ment as a percentage of total employment,
of voters at below-average incomes who stand government expenditure as a percentage of
to benefit from increasing public spending finan- GDP, social security expenditure as a percen-
ced from income taxes. A move towards a more tage of GDP and tax revenue as a percentage
diamond-shaped income distribution pattern of GDP), it does emerge that the two most
lays the conditions for a new tax-conscious slimline governments within OECD (Japan and
winning electoral coalition, and NPM can be Turkey) placed a low degree of emphasis on
represented as the approach to public manage- NPM during the 1980s just as would be
ment which fits this new tax-consciousness in expected. But not all outsize governments in
marginal electorates; for example, by keeping the OECD placed high emphasis on NPM in the
overall public-sector pay bill down by means of 198Os, and the medium-sized governments
performance pay rather than by across-the-board also varied considerably in the emphasis which
pay rises in the traditional style. they laid on NPM. So if government size plays
However, if NPM is best explained as a a part in determining NPM emphasis, it is
response to fiscal stress and government over- probably a subsidiary one rather than the single
load, we might expect NPM to be most strongly determinant.
developed in those countries which score Similarly, the link between macroeconomic
highest on government spending and employ- performance and the degree of emphasis on
ment an&or have a history of relativeIy poor NPM seems to be far from clear-cut. NPM is
macroeconomic performance on the conven- often interpreted as a reflection of 1970s’
tional indices of GDP growth, public debt levels, economic chickens coming home to roost (i.e.
inflation and unemployment rates (such as the record of past or current economic perfor-
Greece or New Zealand). mance), and a return to hard-headed realism.
Indeed, NPM has frequently been interpreted But there appears to be no automatic relation-
(for example, by trade union critics) as little ship between the emphasis piaced by diITerent
more than a means of slimming down big OECD countries on NPM and their level of
government, and saving on resources in the performance on the four conventional macro-
public sector. If slimline public management is economic indicators in the post-oil shock era of
what counts in competition among industrial the 1970s. i.e. 1974-1979.
(or post-industrial) states for economic advan- Table ii gives some indicative data on this
tage, we would expect to see convergence, with point. It is true that some of the macroeconomic
the countries which are least slimline at the success stories of the 1970s are found in the
outset making the most dramatic strides in low NPM emphasis group, as might be expected.
adopting NPM doctrines (because those are the Countries like Japan and pre- 1990 Germany are
countries which would have the most to worry in this group. But not all of the high performers
about in terms of comparative advantage). of the 1970s are in that group, and nor are all
Equally, countries with small-sized public the economic “basket cases” of the 1970s (in
bureaucracy might have proportionately less to term3 of overall scores on GDP growth per head,
gain from putting greater stress on such CPI growth and unemployment) in the high
doctrines. NPM emphasis group. And even if we relate
Finding useful indicators for government size degrees of NPM emphasis to current economic
is no easier than arriving at robust measures of performance (on the same basis) over the
the nature of party political incumbency. Indeed, 198Os, the same puzzles arise, as can also be
it is commonty observed that coventional seen from Table 4. It .seems that macroeconomic
THE NEW PUBLI C MANAGEMENT 103
104
C. HOOD
TABLE 4. NPM emphasis and economic performance 1974-1979 and 1980-1988
Economic
performance
category
Econ era
Hi
Medium Hi
Medium and
Mixed
NPM emphasis 19BOs
Hi Medium lo
1974-1979 19Bo-1988 1974-1979 19Bo-1988 1974-1979 19Elo-1988
JaPm
Sweden Sweden France AUSt l k i BBD BBD
(4)
AUSt i Norway
Japan
(1)
Finland
Norway
N2 (4)
Canada Ireland Greece
AUS Aus
(4)
(4)
Spain
UK (4)
Turkey
(1) (3)
(4)
Medium Lo U.K.
Canada
Lo
IaY
Port.
N2 (1)
‘dY
Port.
(4)
Ireland
(4)
Switz.
Greece
(4)
Spain
(4)
Turkey
(4)
(4) (4)
Source: analysis based on OEW Historical Statistics.
Key to economic performance indicators:
Hi = all available indicators in hi category;
Lo = all avallable indicators ln lo category;
Medium hi = 25% or more of available indicators ln hi category and no indicators in lo ategory;
Medium lo = 25% or more of available indicators in lo category and no indicators ln hi ategory;
Medium and mixed = (a) all available indicators in medium category; (b) available indicators dlsuibuted across all three
categories (hi medium and lo).
Notes:
( 1) No indicator avallable for unemployment rates for this case for this period.
(2) No indicator available for GDP growth per head for this case for this period.
(3) No indicator available for CPl growth for this case for this period.
(4) No indicator available for government debt relative to GDP for this case for this period.
performance alone is not sufficient to explain
the rise of NPM.
As with Table 2, the placings are indicative
and “broad-brush”. NPM scorings are as des-
cribed on p. 101. Economic performance
scorings are based on OECD Historical Statistics
1974-1979 and 1980-1988. Four conventional
series were used: average unemployment
rates as a percentage of total labour force;
average rates of growth in real GDP per
capita; average growth rate in consumer price
index; and average public debt levels as a
percentage of GDP. For each of these indices,
the countries were divided into “hi”, “medium”
and “lo” scorers by the method used as before
(i.e. more than 1 S.D. away from the mean
counted as “hi” or “lo”, and the remainder
“medium”), except that for the CPI index two
outkrs (Turkey.and Iceland) were taken out, in
that both had scores more than twice the mean
and the effect of including them was to put
almost all countries in the “medium” category.
THE NEW PUBLIC MANAGEMENT 105
I nitial endowment
Popular wisdom notwithstanding, there seem
to be important cases which do not readily fit
standard explanations for why NPM developed
in the 1980s. Another possible explanation is
the baseline, or initial endowment from which
different administrative systems start.
Specificially it could be argued that for an
administrative system to move signticiantly
towards NPM, it must be set up at the outset in
such a way as to provide both motive and
opportunity (conventional elements of detec-
tive fiction) for incumbent politicians to want
to shift the administrative system sharply in that
direction.
We could argue that motive in this case might
be expected to consist mainly in the promise
or hope of resource saving from the adoption
of NPM measures, and could therefore be
expected to be proportionately higher in a
context of outsize government and/or acute
fiscal stress associated with poor macroeconomic
performance than in the context of slimline
government and/or strong macroeconomic per-
formance. Opportunity might be expected to
depend on the existence of some “Archimedean
point” from which would-be reforming politi-
cians can influence the public sector as a whole.
For instance, in a country like Switzerland,
where even the number of Ministries (seven) in
the federal government is set out in the constitu-
tion and has not changed for 150 years, the
opportunity for politicians or top officials to
reshape public administration is relatively slight,
because there is effectively no difference bet-
ween constitutional reform and administrative
reform. But in countries like the UK, where
there is no constitutional check to admini-
strative reform and politicians at the centre can
change the entire system, opportunity is much
greater. It would therefore seem that a crucial
variable for opportunity is the extent to which
there is an integrated public service controllable
from a single Point and without significant
jurisdictional breaks (for example, without
independent public bodies beyond the reach of
control by a single set of elected politicians, like
the German Bundesbank).
TABLE 5. Public management baseline styles and propensity
to shift NPM-wards: a tentative hypothesis
Stress on
integration Stress on collectivism in service provision
of public
StTViC~ Lo Hi
Hi
Lo
(1) (2)
“Japanese way’ “Swedish way”
Motive for switch: lo Motive for switch: hi
Opportunity: hi Opportunity: hi
(3) (4)
“American way” “German way”
Motive for switch: lo Motive for switch: hi
00Dommitv: lo Oowrtunitv: lo
Table 5 puts these two aspects of initial
endowment together to identify four polar
types: what is labelled the “Japanese way”,
where public service integration in this sense is
high but collectivism (in the sense of the relative
size of the public sector in spending and employ-
ment) is comparatively low; the “Swedish way”,
where both public service integration and
collectivism are high; the “American way”,
where both integration and collectivism are low;
and the “German way”, where collectivism is
high but integration is low. These labels are
used as convenient shorthand terms, and it
is not suggested that each of these countries
exactly corresponds to the stereotype in all
particulars.
On this basis, some countries would be much
more likely to move NPM-wards than others, at
least in the first round of public management
reforms. For the polar type labelled the “American
way”, there would on these assumptions be
neither motive nor opportunity to make a major
shift NPM-wards. No one would be in a position
to order the changes, and in any case the gains
would be expected to be less than in an outsize
government system. For the “Japanese way”,
there would be opportunity but again little
motive because the system is starting from
the “small government” box. And for the
“German way”, there would be motive, but no
opportunity.”
Only in the polar type labelled the “Swedish
106 C. HOOD
way” w oul d there be both motive and oppor-
tunity: motive, because outsize government
makes resource saving of key importance in
conditions of growing fiscal stress, and oppor-
tunity, because there are central points of
leverage over the entire public service. Hence it
might be argued that countries in the “Swedish
way” box (like the U.K., France, the Scandinavian
countries and possibly the Netherlands) would
be the type most prone to make rapid strides
towards the development of NPM in the
1980s.
A “variable diffusion” model of this kind can
help to explain what more generalized explana-
tions of NPM on their own cannot: namely, why
a number of key OECD countries under
governments of different political stripes shifted
to NPM in the 1980s while others moved
relatively little in that direction. All of the OECD
countries which seem to have placed the highest
stress on replacing PPA with NPM during the
1980s started from the “Swedish way” box in
Table 5 at the baseline, and some of the
countries which put medium-to-high stress on
NPM, such as France and Denmark, also started
as “Swedish way” cases in this sense. Of course,
such an explanation is not independent of
habitat-change accounts stressing the effect of
fiscal stress and tax-consciousness, but it does
explain why a similar habitat change might
produce different effects in different institutional
systems.
CONCLUSION
Compared to the voluminous literature on
deregulation and privatization, accounts of the
rise of NPM are less developed and sparser. We
still lack clear measuring rods for comparing
public management style, and the discussion
here is relatively speculative. Distinguishing
surface change from deep change will always be
difficult in public organizations. However, four
tentative conclusions can be drawn from this
highly exploratory exercise.
First, puce Aucoin, Osborne and Gaebler and
other “global change” interpretations, it is not
clear that the old PPA model of accountability
has collapsed everywhere, or to the same extent.
Even though we do not know how to measure
cross-national differences finely, there do appear
to be leaders and laggards in the process, and it
is interesting that some of the notable laggards
are leading countries in the international eco-
nomy, posing something of a challenge for the
view that public management intemationaliza-
tion parallels economic internationalization. If
a policy dinosaur is going into extinction here,
the process is still far from complete. Hence the
possible relevance of a variable difTusion model
based on initial institutional endowment, as
discussed in the last section. And if the different
institutional endowments lead to permanent
differences (rather than simply governing
the speed of PPA extinction), we may need
to be cautious about assuming that changes
in public sector accounting are likely to be
global.
Second, the conventional explanations of
change in the public sector do not on their own
seem fully to explain observed variations in the
degree to which NPM reforms were taken up by
OECD states in the 1980s. For instance, ideas
such as the view that NPM is all about right-
wingers in office, slimming down outsize govem-
ment or responding to macroeconomic failure
(in the past or present) all come up against
awkward cases which do not seem to fit
expectations when we look across the OECD
countries. But that does not of course rule out
such items as part of a broader multi-factor
explanation of the shift to NPM.
Third, there appears to be no simple relation-
ship between macroeconomic performance
levels and the degree of emphasis laid on NPM.
6 Indeed, Germany itself is an ambiguous case, since its “large government” character derives more from high public
spending than high public employment. If size of public employment is the key tn NPM motivation, there wnuld therefore
be no mare motivation than under the “American way”.
THE NEW PUBLIC MANAGEMENT 107
A working hypothesis might be that the show-
case economies were not under strong pres-
sures to shake up their public management
systems and that the more “basket-case” eco-
nomies lacked the capacity to do so, so that
those making the biggest strides with NPM are
likely to be those in the medium-to-poor
macroeconomic performance bands.
Fourth, there seems to be no simple relation-
ship between the political stripe of governments
(in so far as that can be gauged) and the degree
of emphasis laid on NPM. Are we to assume that
Downs’ (1957) classic ideas about policy con-
vergence in party competition better explain
apparent concensus on NPM than Hibbs’ ( 1977)
ideas about policy divergence? Or could it be
that apparently similar measures have been
adopted in different political circumstances for
diametrically different reasons and with quite
different effects? After all, such things often
happen in administrative reform. Perhaps the
classic historical case is merit hiring for civil
servants, which, according to Hans Mueller
(1984) was adopted by eighteenth-century
Prussia to bring the middle class into the public
bureaucracy and by nineteenth-century Britain
to keep them out.
On similar lines, it might be argued that NPM
has been adopted in some contexts to ward off
the New Right agenda for privatization and
bureaucide and in other countries as the first
step towards realizing that agenda. Much of NPM
is built on the idea (or ideology) of homeostatic
control; that is, the clarification of goals and
missions in advance, and then building the
accountability systems in relation to those
preset goals (cf. Dunsire, 1990). But if NPM has
itself been adopted for diametrically opposite
reasons in ditferent contexts, it may, ironically,
be another example of the common situation in
politics in which it is far easier to settle on
particular measures than on general or basic
objectives. That too may suggest that we should
be cautious about assuming that public-sector
accounting is likely to enter a new age of global
uniformity, at least in the sense of the wider
public management context within which it
operates.
BIBLIOGRAPHY
Aucoin, P., Administrative Reform in Public Management: Paradigms, Principles, Paradoxes and Pendulums,
Gouemunce(l99O)pp. 115-137.
Castles, F. G., Big Government in Weak States: The Paradox of State Size in the English-speaking Nations
of Advanced Capitalism, J ournal of Commonweal th and Comparati ve Politics ( 1989) pp. 267-293.
Castles, F. G. & Merri l l , V., Towards a General Model of Public Policy Outcomes, J ournal of Theoreti cal
Pol ftfcs ( 1989) pp. 177-2 12.
Corbett, D., Pol i ti cs and tbe Ai rl i nes (London: Allen and Unwin, 1965).
Douglas, J., Review Article: The Changing Tide - Some Recent Studies of Thatcherism. Brfti sb J ournal
of Pol i ti cal Scfence (1989) pp. 399-424.
Douglas, M., Cultural Bias, in Douglas, M., In theActi ve Voi ce, chapter 9, pp. 183254 (London: Routledge,
1982).
Dunleavy, P. J,, Bureaucrats, Budgets and the Growth of the State, Bri ti sh J ournal of Pol i ti cal Sci ence
(1985) pp. 29+328.
Dunleavy, P. J., Explaining the Privatization Boom: Public Choice versus Radical Approaches, Publ i c
Admfni stratfon ( 1986) pp. 1334.
Dunleavy, P. J.. Democr acy, Bumaucracy and Publ i c Choi ce (London: Harvester, 1991).
Dunsire, A, Con0vI in a Bureaucracy (Oxford: Martin Robertson, 1978).
Dunsire, A., Holistic Governance, Publ i c Pol i cy and Admi nfstmtfon ( 1990) pp. 3-18.
Dunsire, A. & Hood, C. C., Cutback Management i n Publ i c Bureaucraci es (Cambridge: Cambridge
University Press, 19B9).
Fox, A., Bqond Contract: Work, Power and Trust Rel ati ons (London: Faber and Faber, 1974).
Gorvin, I. (ed.), El ectfons Si nce 1945: A Worl dwi de Reference Compani on (Harlow: Longman, 19B9).
108 C. HOOD
Hall&an, J. & Wettenhall, R, Major Changes in the Structure of Government Institutions, in Power, J.
(ed.), PublicAdministration in Aust?nl&z A Watersbed, pp. 17-40 (Sydney: RAI PAAale and Iremonger,
1990).
Hegewisch, A., Public and Private Sector Trends in Remuneration Policies and Contract Fleslbility in
Europe, Paper presented at seminar International Come of Public SectorPay I n 10 Countries~
1990 Results j?vrn tbe Price Waterbouse0anfield Survey (London: Public Finance Foundation,
5 September 199 1).
Hibbs, D. A. Jr, Political Parties and Macroeconomic Policy, Am&curt Pollticul Science Reuiew (1977)
pp. 1467-1487.
Hirschman, A. O., Shifting I nuolwmen ts: Mu&e I nterest and Public Action (Oxford: Blackweli, 1982).
Hood, C. C., Public Administration, in Bogdanor, V. (ed), Fbe Blackwell Encyclopaedia of Poll&al
I nstitutions, pp. 504-507 (oxford: Blackwell, 1987).
Hood, C. C., Public Administration: lost an Empire, Not Yet Found a Role, in Leftwich, h (ed.), New
Ditvcttons in Polftical Scfence, pp. 107-125 (Aldershot: Elgar, 19901).
Hood, C. C., Beyond the Public Bureaucracy State? Public Administration in the 1990s (Inaugural lecture,
London School of Economics, 16 January 199Ob).
Hood, C. C., De-Sir-Humphrey-tying the Westminster Model of Governance, Go uemunce (199Oc) pp.
205-214.
Hood, C. C., A Public Management for Ail Seasons?, Public Administration (1991) pp. 3-19.
Hood, C. C. & Dunsire, A., Bureaumetrics (Famborough: Gower, 1981).
Hume, L. J., Bentham and Bureaucracy (Cambridge: Cambridge University Press, 1981).
Kamenka, E., Bureaucracy (Oxford: Blackwell, 1989).
Karl, B. D., fiecutiw Reorganfzatfon and Reform in tbe New Deal (Cambridge, MA: Harvard University
Press, 1963).
Marini, F., Towurd a New Public Administrution (Scranton, PA: Chandler, 1971).
Martin, J., A Profession of StatecraJ ?? Tbrtv J &says on Some Cumsnt hues in the New Zealand Public
Service (Wellington: Victoria University Press, 1988).
OECD, Public Management Developments (Paris: OECD, 1988).
OECD, Public Management Dewlopments (Pa& OECD, 1990).
Osborne. A. & Gaebler, T., Reinwnting Government: How tbe Enhqreneu~l Spfrft is Transf~ing
tbe Public Sector (Reading MA: Addison-Wesley, 1992).
Painter, M.. Values in the History of Public Administration, in Power, J. (ed.), Public Administrution I n
Australia: A Watersbed, pp. 75-93 (Sydney: RAIPAHale and Iremonger, 1990).
Peters, 8. G. & Heisler, M. 0.. Thinking About Public Sector Growth: Conceptual, Operational, Theoretical
and Policy Considerations, in Taylor, C. L (ed.), Wby Gowmmenfs Grow Measuring Public Sector
Size, pp. 177-197 (Beverly Hills: Sage, 1983).
Piore, M. J. & Sahel, C. F., The Second J ndxHrhal Divide (New York: Basic, 1984).
Pollitt, C., Managerialism and tbe Publk Services Tbe Anglo-American Ekperhce, 2nd Edn (Oxford:
Blackwell, 1993).
Power, M. & laughlin, R., Critical Theory and Accounting in Alveson, N. & Willmott, H. (eds), Critical
Management Studies, pp. 115135 (London: Sage, 1992).
Pusey, M.. Economic Rationalism in Canbewa (Cambridge: Cambrklge University Press, 1991).
Quirk, P., In Defense of the Politics of Ideas,jouml of Politfa (1988) pp. 31-41.
Schaipf, F. W.. The Political Calculus of Inllatlon and Unemployment in Weatem Europe: A Game-Theoretical
Interpretation, J ournal of Public Policy (1987) pp. 227-258.
Scott. G.. Bushnell, P. & Sallee, N.. Reform of the Core Public Sector: New Zealand Experience, Gouemunce
(1990) pp. 138-167.
Sjolund, M., Statens Liinepolitik 1966-19&l (Stockholm: Allmanna Fiirlaget, 1989).
Small, A. W., The Camerulfsts (Chicago: Chicago University Press, 1909).
Sturgess. G.. First Keynote Address, in Carey, G. & Ryan, P. (eds), In Tnansttiotz NSW and t&e
Corporatisatlon Agenda, pp. 4-10 (Sydney: Macquarie Public Sector Studies Progmm/Association for
Management Education and Research, 1989).
Taylor, J. A., Information Networking in Public Administration, Intern&o MI Review of Admlnistrat&e
Sciences (1992) pp. 375-389.
Taylor, J. A. & Williams, H., Public Administration and the Information Polity, Public Admdnfshation
(1991) pp. 171-190.
THE NEW PUBLlC MANAGEMENT
109
Thompson, M., Ellis, R & Wilds&y, A., Cultural Tbtwy (Boulder: Westview, 1990).
von Beyme, IL, Do Parties Matter? The Impact of Parties on the Key Decisions in the Political System,
Government and Op&nwition (1984) pp. 529.
Yeatman, A_, ‘Ihe Concept of Public Management and the Australian State in the 198Ck,Au.strulfun J ournul
of Publ i c Admi ni strati on (1987) pp. 339-353.
doc_390487559.pdf
Changes in public sector accow~tingin a number of OECD countries over the 198Bs were central to the
rise of the “New Public Mulaganent” (NPM) and its asso&md doctrks of public accoumabihty and
organizitiooal best practice. lhis paper d&uases theriseofNPMuulalvmativetothetnditionofpuMic
accountabilityembodiedin pmgmssive-era public admin&ra&n ideas. It argues that, in spite of allegations
ofintem?tionalintionurdtheadoptionofanewgiobalp;udigminpuMic mylaganent,d==w=
considerable variation in the extent to which different OECD countries xlopted
Ac an#i n& ol gaf uul l ons and Soc i et y, Vol . 20, No. U3, pp. 93-109, 1995
Ekvicr Science Ltd
Printed in Great Britain
0361~36Bzl95 19.50+0.00
THE “NEW PUBLIC MANAGEMENT” IN THE 1980s:
VARIATIONS ON A THEME’
CHRISTOPHER HOOD
London School of Economics and Political Science
Changes in public sector accow~ting in a number of OECD countries over the 198Bs were central to the
rise of the “New Public Mulaganent” (NPM) and its asso&md doctrks of public accoumabihty and
organizitiooal best practice. lhis paper d&uases theriseofNPMuulalvmativetothetnditionofpuMic
accountabilityembodiedin pmgmssive-era public admin&ra&n ideas. It argues that, in spite of allegations
ofintem?tionalintionurdtheadoptionofanewgiobalp;udigminpuMic
mylaganent,d==w=
considerable variation in the extent to which different OECD countries xlopted NPM over the 19t3Os. It
further argues that conventiortal expMm&u of the rise of NPM (“w, party pohticai incumbency,
etonomic~tecorduldgovanment~)seanhudtosust?inevcn6romarrl?tivclybrief
inspectionofsuchaossnationaldua~ucavailable,uldthatulaplul?tionbvedoninitialendowmcnt
maygiveusaditkrentpeqectiveonthosechanges.
Over the 1980s there was a move in a number
of OECD countries towards the New Public
Management (NPM). Central to this change in
modes of public management was a shift towards
“accountingization” (a term coined by Power 81
Laughlin, 1992, p. 133).’ This development can
be claimed to be part of a broader shift in
received doctrines of public accountability and
public administration At the same time, account-
ing changes formed an important part of the
assault on the progressive-era models of public
accountability (cf Hal&an & Wettenhall, 1990).
For progressive public administration,2 demo-
cratic accountability depends on limiting corrup-
tion and the waste and incompetence that are
held to go with it (cf. Karl, 1963, p. 18). The
assumption is that politicians are inherently
venal, using their public office wherever pos-
sible to enrich themselves, their friends and
relations, and that reliance on private-sector
contracting for public services inevitably leads
to high-cost lowquality products, either because
of corrupt influence on the contract-awarding
process or because the public contract market
will come to be controlled by organized crime,
or both. Whether these assumptions can be
safely dispensed with in the wealthy OECD
countries of today is a matter for debate.
From those assumptions, the accountability
paradigm of progressive public administration
(hereafter PPA for convenience) put heavy
stress on two basic management doctrines. One
of those doctrines was to keep the public sector
sharply distinct from the private sector in terms
l This paper was presented in an earlier form to the Workshop on “Changing Notions of Accountability in the UK Public
Sector”, SE, December 1991, and to the ElASM Workshop on Accounting, Accountability and the “New European Public
Sector”, Helsinki, September 1992.1 am grateful for comments and criticisms received on those occasions, and particularly
to Peter Miller for helpful suggestions.
’ Broadly, “accountingization” means the introduction of ever-more explicit cost categorization into areas where costs
were previously aggregated, pooled or undefined.
* That is, the style of public administration that emerged in the “progressive era” of the late nineteenth and early twentieth
centuries.
93
94
C. HOOD
of continuity, ethos, methods of doing business,
organizational design, people, rewards and
career structure. The aim, in Beatrice Webb’s
words (Barker, 1984, p. 34) was for a “Jesuitical
corps” of ascetic zealots. The other doctrine was
to maintain buffers against political and mana-
gerial discretion by means of an elaborate
structure of procedural rules designed to
prevent favouritism and corruption and to keep
arms-length relations between politicians and
the entrenched custodians of particular public
service “trusts”.3
This organizational model attracts more deri-
sion than analysis today (cf. Osborne & Gaebler,
1992). In fact, it reflects an underlying metaphor
of trustee and beneficiary (John Locke’s meta-
phor for government) and involves a complex
mix of high-trust and low-trust relationships, with
the accompanying accounting rules reflecting
degrees of trust. Within the “Jesuitical corps” of
the public service were many high-trust rela-
tionships (for example, in conventions of mutual
consultation or action on the basis of word-of-
mouth agreements across departments), the
costs of which were not accountingized. The
implicit assumption is that such high-trust,
non-costed behaviour lowers transaction costs
within the public sector and makes it more
efficient than it would be if each action had to
be negotiated and costed on a low-trust basis.
However, PPA also embodied many low-trust
relationships, particularly in areas where the
Jesuitical corps faced the corrupting forces of
the world outside, notably the award of con-
tracts, recruitment and staffing, as well as the
handling of cash, where distrust prevailed and
elaborate records had to be kept and audited.
In the place of the PPA model came New
Public Management or NPM (cf. Aucoin, 1990;
Hood, 1987, 1990a, b, 1991; Dunsire & Hood,
1989; Hood & Jackson, 1991; Pollitt, 1993;
Pusey, 199 1). NPM involved a different concep-
tion of public accountability, with different
patterns of trust and distrust and hence a
different style of accountingization. The basis of
NPM lay in reversing the two cardinal doctrines
of PPA; that is, lessening or removing differences
between the public and the private sector and
shifting the emphasis from process account-
ability towards a greater element of account-
ability in terms of results. Accounting was to be
a key element in this new conception of
accountability, since it reflected high trust in
the market and private business methods (no
longer to be equated with organized crime) and
low trust in public servants and professionals
(now seen as budget-maximizing bureaucrats
rather than Jesuitical ascetics), whose activities
therefore needed to be more closely costed and
evaluated by accounting techniques. The ideas
of NPM were couched in the language of
economic rationalism, and promoted by a new
generation of “econocrats” and “accountocrats”
in high public office.
The term NPM was coined because some
generic label seemed to be needed for a general,
though certainly not universal, shift in public
management styles. The term was intended to
cut across the particular language of individual
projects or countries (such as the French ‘Projet
de Service”, the British “Next Steps”, the
Canadian “Public Service 2000”). The analogy
is with terms like new politics, new right, and
new industrial state, which were invented for a
similar reason.4
As with the disappearance of the dinosaurs,
there is no single accepted explanation of this
alleged paradigm shift. In fact, emerging expla-
nations roughly parallel the major contending
theories of the dinosaurs’ extinction. Some
.3 An organizational structure which could clearly be classed as “hierarchist” in the cultural theory of Mary Douglas ( 1982)
and her followers (Thompson et al.. 1990).
4 The term “new” does not imply that NPM doctrines appeared for the first time in the 1980s (any more than the “New
Learning” of the fifteenth and sixteenth centuries involved the first discovery of Latin and Greek). Many NPM doctrines
repackage ideas which have been in public administration since its earliest beginnings. Nor must NPM be confused with
the “New Public Administration” movement in the U.S.A. in the late 1960s and early 1970s. which achieved no real
mainstream influence (see Marini, 1971).
THE NEW PUBLIC MANAGEMENT 95
accounts stress “habitat loss” for the old style
arising from post-industrial technology, pro-
ducing a wholly new model of public admini-
stration built around electronic data handling
and networking, providing many new niches for
accountingization and lowering its direct costs
(see Taylor & Williams, 1991, p. 172; Taylor,
1992). Some see the demise of the old model
as the result of a sudden shock, with New Bight
ideas about organizational design coming as a
meteorite from out of the blue (as in Quirks,
1988, idea-centric account of 1980s’ deregula-
tion). Some see PPA’s fate as a self-induced
extinction, as older control frameworks and
accounting practices came to degrade the values
which they were designed to promote (see
Painter, 1990, p. 77; Hirschman, 1982). And yet
others interpret the change as caused by a new
set of predator interests, such as accounting
firms and management consultants, hunting PPA
into extinction (see Dunleavy, 1985, 1986,
1991).
contains important variations, and that no
account of the shift from the progressive public
administration model to NPM can be satisfactory
unless it can account for international leaders
and laggards.
NEW PUBLIC MANAGEMENT; THEMES
AND VARIATIONS
Themes
However, before such “extinction science”
can be developed, we need to be satisfied that
some general extinction has actually taken
place, and that the new life-form of NPM is
everywhere supplanting PPA. Such claims are
indeed commonly made both by practitioners
and by academic commentators. Aucoin (1990,
p. 134) for instance, asserts that: “What has been
taking place in almost every government in
developed political systems and highly institu-
tionalized administrative states is a new emphasis
on the organizational designs for public manage-
ment . . This internationalization of public
management parallels the internationalization of
public and private sector economies.” Similarly,
Osborne & Gaebler (1992, pp. 322-330) write
of NPM as a new “global paradigm”, claiming
that transition to the new paradigm is inevitable
“just as the transition from machine rule to
Progressive government was inevitable” (p.
325).
The doctrines of public sector management
encompassed by NPM have been variously
described by different commentators (such as
Aucoin, 1990; Hood, 1991; Pollitt, 1993) and
some have identiBed diRerent phases in the
development of NPM. However, there is still a
good deal of overlap among the different
accounts of what NPM entailed. For example,
the idea of a shift in emphasis from policy making
to management skills, from a stress on process
to a stress on output, from orderly hierarchies
to an intendedly more competitive basis for
providing public services, from fared to variable
pay and from a uniform and inclusive public
service to a variant structure with more
emphasis on contract provision, are themes
which appear in most accounts.
Most commentators have associated NPM
with approximately seven dimensions of change,
which are summarized in Table 1, together with
their associated doctrines, and some speculative
ideas about their implications for accountingiza-
tion. The elements relate to the two cardinal
elements of PPA already noted, in that the first
four elements of Table 1 relate to the issue of
how far the public sector should be distinct from
the private sector in its organization and
methods of accountability, and the last three
broadly relate to the issue of how far managerial
and professional discretion should be fenced in
by explicit standards and rules. The seven
elements are as follows.
If we accept such claims, we would expect ( 1) A shift towards greater disuggmgution of
to see a process of international convergence public organizations into separately managed
and diffusion of NPM ideas in public administra- “corporatized” units for each public sector
tion. But this paper argues that the inter- “product” (each identified as a separate cost
nationalization of the NPM model at least centre, with its own organizational identity in
% C. HOOD
TABLE 1. Doctrinal components of new public vt
No. Doctrine T)piCd Replaces OpentiOMl Some possible
justification significance accounting
imDliatiOnS
PS dtsftncttwness
Unbundling of the
Ps into corporatized
units organized by
product
More contract-
based competitive
provision, with
internal markets and
term contracts
Stress on private-
sector styles of
management
practice
More stress on
discipline and
liugality in resource
USe
Rules us dtscwtton
5. More emphasis on
visible hands-on top
management
6. Explicit formal Accountability
measurable stan- means clearly stated
dards and measures aims; efficiency
of pefonnce and needs hard look at
success goals
7. Greater emphasis
on output controls
Need for greater
stress on results
Make units
manageable, and
focus blame; split
provision and
production to create
anti-waste lobby
Rivahy as the key to
lower costs and
better standards;
contracts as the key
to explicating
performance
SGMdd.3
Need to apply
proven private-
sector management
tools in the public
sector
Need to cut direct
costs, raise labour
discipline, do more
with less
Accountability
requires clear
assignment of
responsiblity not
difision of power
Belief in uniform Erosion of single
and inclusive PS to service
avoid underlaps and employment; atms-
overlaps in length dealings;
accountability devolved budgets
Unspecified
employment
contracts, open-
ended provision,
linking of purchase,
provision,
production, to cut
transaction cost
Stress on PS ethic
fixed pay and hiring
rules, model
employer
orientation
centralized
personnel structure,
jobs for life
Stable base budget
and establishment
norms, minimum
standards, union
vetoes
Paramount stress
on policy skiUs and
rules, not active
management
Qualitative and
implicit standards
and norms
Stress on procedure Resources and pay
and control by based on
collibration performance
activity
Source: adapted from Hood (1991, pp. 4-5).
More cost centre units
Distinction of
primvy and
secondary public
service labour force
Move from double
imbalance PS pay,
career service,
unmonetized
rewards “due
process” employee
entitlements
Less Primary
employmen& less
job security, less
producer-friendly
style
More stress on
identifying costs and
understanding cost
structures; so cost
data become
commercially
confidential and
cooperative behaviour
becomes costly
Private-sector
accounting norms
More stress on the
bottom line
More freedom to
manage by
disaetionaty power
Erosion of self-
management by
professionals
Fewer general
procedural constraints
on handling of
contracts, cash, staff;
coupled with more use
of fmancial data for
management
accountability
Pertixmance
indicators and audit
Move away from
detailed accounting
for particular activities
towards broader cost
centre accounting;
may involve blurring
oftundsbrpayandfor
THE NEW PUBLIC MANAGEMENT 97
fact if not in law, and greater delegation of
resource decisions, in a movement towards
“one-line” budgets, mission statements, business
plans and managerial autonomy). The corpora-
tized style contrasts with the PPA style of
providing all public services through “semi-
anonymized” organizations within a single
aggregated unit, with detailed service-wide
rules, common service provision in key areas of
operation, detailed central control of pay
bargaining and stafhng levels.
(2) A shift towards greater competition both
between public sector organizations and bet-
ween public sector organizations and the private
sector. The aim for a more competitive style
contrasts with the PPA style of ascribing semi-
permanent “ascribed” roles to public sector
organizations; that is, captive markets which are
indefinitely assigned to particular “prestige”
producers.
(3) A move towards greater use within the
public sector of management practices which
are broadly drawn from the private corporate
sector, rather than PPA-style public-sector-
specific methods of doing business. Examples of
the latter include “model employer” aspirations
to set an example to, rather than to follow the
lead of, private-sector employers in matters of
pay and conditions of employment (for example,
in equal opportunity or employment of disabled
persons) and the traditional “double imbalance”
pay structure of public administration, in which
lower-level statf tend to be relatively highly paid
compared to their private-sector counterparts
and top-level staff are relatively low-paid (cf.
Sjolund, 1989).
(4) A move towards greater stress on disci-
pline and parsimony in resource use and on
active search for finding alternative, less costly
ways to deliver public services, instead of laying
the emphasis on institutional continuity, the
maintenance of public services which are stable
in “volume terms” and on policy development.
( 5) A move towards more “‘hunds-on manage-
ment” (that is, more active control of public
organizations by visible top managers wielding
discretionary power) as against the traditional
PPA style of “handsofF management in the
public sector, involving relatively anonymous
bureaucrats at the top of public-sector organiza-
tions, carefully fenced in by personnel manage-
ment rules designed to prevent favouritism and
harassment.
(6) A move towards more explicit and
measurable (or at least checkable) stanahrds
of performance for public sector organizations,
in terms of the range, level and content of
services to be provided, as against trust in
professional standards and expertise across the
public sector. The old PPA style involved low
trust in politicians and managers but relatively
high trust in professional expertise, both in a
“vertical” sense (that is, up and down the
organizational ladder, or between “principals”
and “agents” in the new legal+conomic lan-
guage of the economic rationalists) and in a
“lateral” sense (that is, across difyerent units of
the public sector; cf. Fox, 1974, pp. 72-84,
102-l 19).
(7) Attempts to control public organizations
in a more “homeostatic” style according to pre-
set output measures (particularly in pay based
on job performance rather than rank or educa-
tional attainment), rather than by the traditional
style of “orders of the day” coming on an ad
hoc basis from the top, or by the subtle balancing
of incompatible desiderata in the “collibration”
style of control identified by Dunsire (1978,
1990) as central to orthodox bureaucratic
functioning.
These doctrines of NPM link to recurrent
debates about how public administration should
be conducted, which stretch back at least as far
as the major disputes between “legalists” and
“Confucians” in the Chinese mandarinate over
2000 years ago (see Kamenka, 1989, pp.
38-39). How far the public sector should be
insulated and clearly separated from the private
sector in matters of handling business and staff,
and how far business should be conducted by
professional discretion rather than by pre-set
rules or standards, are issues which go to the
heart of most doctrinal disputes in public admini-
stration, including such major waves of classic
public administration thought as the ideas of the
German cameralists from the mid-sixteenth
98
C. HOOD
century (Small, 1909) the nineteenth-century
British utilitarians (Hume, 1981) and the turn-
of-the-century American progressives (Ostrom,
1974). Such doctrines also have profound
implications for how public sector accounting
is conceived, in the sense of what records are
kept, how they are used, and what is costed and
measured.
This summary list of course oversimplifies,
and there are many interesting counter-trends.
Examples of such counter-trends include: the
unfashionability of the traditional public enter-
prise model in conventional market sectors of
the economy, coupled with the vigorous adop-
tion of that model for non-marketed public
services in several countries; the weakening of
older doctrines of metaphytic competition (i.e.
public versus private providers; Corbett, 1965)
as against doctrines of market testing by
franchising; the weakening of trust in profes-
sionals while strengthening the hand of mana-
gers. Certainly, there is no logical necessity for
a public management system to change in all of
these seven respects at once. Many variations
are possible.
Variations
There are no systematic cross-national studies
showing degrees of variation in public manage-
ment reform in a robust and reliable way. The
literature in the area is long in anecdote and
general commentary but short on systematic
comparison, and comes close to being a data-
free environment. There are only isolated
fragments and relatively low-grade comparative
data from sources like OECD public manage-
ment reports and cross-national consultancy
reports such as the Price WaterhouseKranfield
study of comparative pay flexibility (Hegewisch,
1991).
But even such fragmentary sources are suffi-
cient to show that not all OECD countries moved
to adopt NPM principles to the same extent during
the 1980% and that there were marked diffe-
rences even within similar family groups such
as the English-speaking “Westminster-model”
countries (cf. Hood, 1990~). It is particularly
notable that some of the OECD’s showcase
economies, Japan, Germany and Switzerland,
seem to have put much less emphasis on
adopting NPM-type reforms (on the seven
dimensions indicated in Table 1) in the 1980s
than countries like Sweden, New Zealand or the
U.K. But it would be hard to argue that those
countries were closest to the NPM model at the
outset, particularly in respect of use of private
sector-style management practices, hands-on
management or output controls.
For example, the NPM tendency to decentra-
lize personnel management (such as hiring and
job classifications) to operating units away from
central oversight agencies was not a marked
tendency in Japan, where the National Personnel
Authority was if anything strengthened rather
than weakened over the 1980s. Administrative
reform received much attention, in the form of
the three reform commissions over that decade
(modelled on the famous 1937 U.S. “Brownlow”
Committee on Administrative Management),
but the accent seems to have been more on
privatization, deregulation and tax reform than
on the principles of NPM, with the exception
of a small measure of corporatization in the form
of more freedom for Ministries to reorganize
themselves without specific authority from the
Diet. And whereas doctrines of “pay for perfor-
mance” took a strong hold in countries such as
Sweden, Denmark, New Zealand and the U.K.,
there was no equivalent movement in Germany
(partly because pay for performance poten-
tially conflicts with the Basic Law doctrine of
equality of pay across particular grades in the
public service). Indeed, neither Germany nor
Switzerland made any major changes in their
public administration at federal level in the
1980s; rather, Vewaltungspjlege (a quiet period
of cultivation after the reforms of the 1960s which
embraced policy evaluation) was a common
watchword in Germany over that decade.
But even within the group of countries in
which more emphasis seemed to be placed on
public management reforms, it is not clear that
the direction of change was the same. For
instance, the “free commune” experiment in
Norway in 1987 and the French move to far-
reaching territorial decentralization in 1983
THE NEW PUBLIC MANAGEMENT 99
contrasts sharply with centralizing tendencies
in the U.K. over the 1980s. Late 1980s reforms
in the U.K. and New Zealand were aimed at
separating policy setting and service provi-
sion (on the grounds that Ministers were not
equipped to be managers), while the Australian
Commonwealth government took measures
intended to strengthen the capacity of Ministers
to manage.
Any scoring of variations in the adoption of
NPM in the seven dimensions discussed above
must necessarily be highly impressionistic.
Ideally, we would need both a reliable method
of locating a country’s initial state at the start
of the period in question and the extent of
movement over the period (given that a
“backward” case might show dramatic change,
and yet still be behind an apparently “static”
country which started from a higher initial
emphasis on the NPM style). At present, we do
not have a cross-national information source
which could reliably show either sort of
variation.
From the fragmentary literature on public
management reform over the 1980s the high
NPM group in the OECD countries would be
likely to include Sweden, Canada, New Zealand,
Australia and the U.K. with France, Denmark,
the Netherlands, Norway and Ireland also
showing a number of marked shifts in the
direction of NPM. At the other end, the low NPM
group would be likely to include Germany,
Greece, Spain, Switzerland, Japan and Turkey.
These impressions are consistent with the
country reports submitted to the OECD’s survey
of public management developments in 1988
and 1990, supplemented in 1991 (OECD 1988,
1990, 1991). Obviously, these reports are
seriously contaminated in a number of familiar
ways, mainly because they reflect what the
correspondents used by OECD in each country
thought it relevant or politic to record (although
there was, of course, a general check-list issued
by PUMA for these exercises), rather than what
a single overall observer might have noted.
Moreover (until such time as the OECD’s public
management profiles are greatly developed), it
indicates what was on the government agenda
for change rather than the “initial endowment”
of each system.
Even so, OECD public management reports
can give us a rough indication of the officially
perceived agenda of public management change
in the 198Os, and serve as a starting point for
discussion of variations. The country reports for
1988 and 1990 were carefully examined, and a
rough score awarded to each country under
each of the seven points of public management
doctrine which were mentioned above as
components of new public management. A score
of 2 was given for developments reported as
being in place on each of the dimensions, 1 for
developments under active discussion or experi-
mentation and 0 for nothing reported in the
area.
Clearly, such an exercise is only useful for
identifying outliers, and not for making any fine
difTerentiations. The extremes were taken as
cases whose overall scores (summed across all
seven dimensions) were more than one standard
deviation away from the mean. The high and
low cases identified in this way fitted sufficiently
well with other impressionistic views of varia-
tion in NPM over the 1980s to serve as a basis
for a discussion as to what might be responsible
for putting a country at one or other of the
extremes.
WHY THE VARIATIONS?
As noted earlier, there are several different
possible ways of explaining the rise of NPM. This
section explores four conventional accounts of
public sector change (relating such change to
“Englishness”, party politics, government size
and macroeconomic performance) before sug-
gesting a further interpretation couched in
terms of institutional endowments.
‘English awfulness”?
Some commentators (notably Pollitt, 1993)
have implied that NPM was mainly an Angle+
American phenomenon of the ReaganThatcher
era. But this view seems difficult to sustain. If
nothing else. it ignores the high degree of
100
C. HOOD
emphasis placed on NPM in South Africa, Hong
Kong, Australia and New Zealand (for the latter,
see Pusey, 1991; Scott et al., 1990; Yeatman,
1987).
However, it might be possible to broaden tire
view of NPM as an Anglo-American preoccupa-
tion to the idea that it reflects what Castles
ironically calls “the awfulness of the English”.
By this phrase, Castles means the relatively
poor economic performance and arrested deve-
lopment of welfare state policies which, he
claims, characterize the English-speaking coun-
tries (Castles, 1989; Castles & Merrill, 1989, pp.
181-185). It is, for example, noticeable that the
high scorers on NPM emphasis are mostly English-
speaking countries (and hence clearly candidates
for “English awfulness” ti Zu Castles). The low
scorers, in contrast, are all non-English-speaking
countries.
Moreover, an “English awfulness” explanation
might fit with a view of NPM as representing
international convergence on a common public
management style. On Castles’ analysis, the
English-speaking countries lost their formerly
distinctive “high-direct employment” feature of
public management between the 1960s and the
198Os, and NPM could be interpreted as part of
that process of coming into line with non-
English-speaking countries. Particularly with
respect to the practice of accounting tech-
niques and management consultancy, diflu-
sion of the new NPM ideas might have been
expected to spread more readily across coun-
tries with the same language and similar legal
traditions.
But it seems too simple to attribute NPM to
English awfulness alone. For example, Sweden
appears as a high scorer on NPM emphasis, which
would be particularly damaging for an interpre-
tation built on a Castles-type English awfulness
factor. Moreover, Denmark, the Netherlands and
France are also cases which score relatively high
on NPM emphasis, and two of them (Denmark
and the Netherlands) reported strong develop-
TABLE 2. NPM emphasis and political incumbency
Political incumbency emphasis
NPM
Emphasis “LeW “Centre”
” WW
Hi Sweden
Medium
Lo
France
Greece
NZ
Austria
Denmark
Finland
ItafY
Netherlands
Portugal
USA5
Germany
JaPm
Snain Switzerland Turkev
Sources: analysis bved on OECD PUMA reports and on
political incumbency data for OECD countries drawn from
Gorvin (1989) and Keestng’s Contem~ Ar c hf ves.
ment of variable pay in recent years to the 1990
Price WaterhouseICranfield project (Hegewisch,
199 1, Table 3). NPM seems to be more than just
another “English disease”.
Party politics
Some commentators explain PPA’s demise in
rather similar terms to the predator theorists of
the dinosaurs’ extinction. The notion is that the
old structure has been subverted by the deve-
lopment of “New Right” interests who stand to
benefit in various ways from dismantling the
PPA model.
At one level, the demise of traditional PPA is
often attributed to the advent of New Right
government in the 1980s and particularly to
the infiuence of Ronald Reagan and Margaret
Thatcher, who aimed to roll back big govern-
ment and state-led egalitarianism and welfarism,
and to remould what was left of the public sector
in the image of private business (see Pollitt,
1993). lf NPM was sparked by such figures, we
would expect its development to be most
marked in countries which were governed by
right-wing parties during the 1980s.
’ It is obviously debatable whether U.S.A. should be scored as “centre” or “right” for these purposes. In many “does party
matter” studies. it is excluded, on the grounds that it is essentially unclassifiable. Given its separation of powers, it is here
counted asacentre-leftcoalitionfor 1980,acentre-right coalition for 1981 to 1986andacentristcoalitionfor 1987 to 1990.
THE NEW PUBLIC MANAGEMENT 101
Table 2 gives a rough indication of the extent
to which OECD countries were governed during
the 1980s mainly by parties on their left, their
right or their centre. For this exercise, a score
of i- 1 was given to each OECD country for each
year of incumbency in government by a political
party to the right of that country’s political
spectrum, and a score of -1 for each year of
incumbency by a party to the left of that
spectrum; 0.5 was given for each year of
incumbency by a centre-right coalition and
-0.5 for each year of incumbency by a centre-
left coalition, with 0 for a grand coalition (as in
the Austrian case). Countries which score hi
(more than 1 S.D. above the mean) are taken as
“right” in incumbency terms, and countries
which score lo (more than 1 S.D. below the
mean) are taken as “left” in incumbency terms.
“Centre” scorers are the rest.
NPM scorings are based on the country
reports to OECD on Public Management (1988
and 1990), with a maximum, score of 2 on each
of seven dimensions of doctrine, as described
earlier. “Hi” scorers are those whose overall
score is more than 1 standard deviation (S.D.)
above the mean score; “lo” scorers are those
whose overall score is more than 1 S.D. below
the mean score. “Medium” scorers are the
rest.
This exercise is useful only for first approxi-
mations, and it may be that a better test would
be a series of “before” and “after” looks at cases
of changes in government. Scoring political
incumbency involves problems which are fami-
liar in the “does politics matter?” debate over
macroeconomic policy in political science.
There is no established method for comparing
degrees of “rightness” and “leftness” ac?wss
countries, and there are clearly dilYerent “qual i -
ti es” of rightness and leftness (for example,
participatory versus hierarchical emphases in
socialism). Presidential and federal systems
clearly cause complications (for example, the
I7.S. would count as “right” only in Presidential
terms for the 1980s). Coalition/PR system cases
tend to bunch in the middle, so that the outliers
rend to be non-PR systems like France and Japan,
which may get disproportionate weight in the
analysis. All that such a scoring exercise does is
to give us a first cut as to whether OECD
countries were governed during the 1980s
mainly by parties on their left, their right or
their centre.
Crude as the data are, however, Table 2 shows
up the difiiculties in the popular idea that NPM
was closely associated with incumbency by
“right” governing parties in the 1980s. Sweden
is the most obvious misfit for such an idea. It is
a country which shows apparently high NPM
emphasis during the 198Os, but also scores fairly
high for left political incumbency with eight
years out of the decade under Social Democratic
governments. Indeed, Sweden is conventionally
taken as the leading case of the social demo-
cratic alternative to liberal capitalism. And at
the other extreme, there are unambiguously
“right” cases, like Japan and Turkey, which seem
to score distinctly low on the NPM emphasis
scale.
Of course, such results are only surprising if
we expect the incumbency of dserent political
parties to lead to different public policy
outcomes. Other analyses of party competition
might lead to different expectations. An example
is Scharpfs ( 1987) “nested game” model of the
dynamics of party competition, in which two
rival “left” and “right” parties compete for the
fickle favours of floating middle-ground voters
against a set of variant macroeconomic condi-
tions produced by a game between incumbent
governments, labour unions and central banks.
Modifying that framework only slightly, we
could posit a “party competition” game played
between right and left wing parties (ri la
Scharpf) and a “public management game”
played between incumbent governments and
public managers/top bureaucrats or profes-
sionals (substituted for Scharpfs macroeconomic
policy game played between governments and
labour unions). In the second game, incumbent
politicians choose between tough and tender
approaches to public managers, and the latter
choose between a cooperative and uncoopera-
tive approach to politicians. Table 3 outlines
such a game. Clearly, politicians would prefer
to be in cell ( l), bureaucrats and managers in
102 C. HOOD
TABLE 3. Politician-bureaucrat public management game
Bureaucrats/
managers/
Politicians
public servants Tender Tough
Cooperative
(1) (2)
Outcome: smooth Outcome:
running medium cost politicians
public management exploit public
service ethos for
cheap and
effective public
management
Noncooperative
(3) (4)
Outcome: Outcome: high cost
bureaucrats exploit administrative
politicians’ chaos
goodwill for high
cost public
management
cell (3); such a game therefore has a tragic bias
away from cell ( 1) and towards cell (4).
If such a model of party competition is
adopted, it becomes quite conceivable that
incumbent social-democratic governments (such
as those in Sweden or New Zealand in the 1980)
might go just as far as “bourgeois” governments
(and possibly even further) in moving towards a
tough approach to public management, because
they have nowhere else to go, and need to work
harder to establish credibility in this area with
wavering middle-class voters (adapting Scharpfs
ideas of asymmetrical choice by the middle
socioeconomic group).
From such evidence, it would seem that if the
rise of the New Right influenced the demise of
the old PPA model, it was not simply through
right-wing parties holding office. It looks more
like a general process of policy diffusion than a
process in which policies vary with party
incumbency. A modified version of the rise of
the New Right explanation would therefore
need to place more stress on the eclipsing of
the old public administration coalition (parti-
cularly the public service labour unions) by an
NPM coalition drawn from accounting firms,
financial intermediaries, management consul-
tants and business schools, along the lines of
Patrick Dunleavy’s (1986) account of how
policy booms develop through a coalition of
professional and corporate interests.
Alternatively (perhaps additionally), the change
might be attributed to the recolonization of the
public service from inside rather than to an
assault from outside. Some contemporary com-
mentators on the rise of NPM, notably Yeatman
( 1987, pp. 350-351) and Pusey ( 1991) have
applied a form of “new class” analysis to the
process, in the tradition of Bumham’s (1942)
analysis of managers as a new class. They claim
that the upper echelons of the public service
are increasingly being occupied by a new class
of “econocrats” (and perhaps “accountocrats”,
certainly in the New Zealand case), who were
not fired up into public service ideals by close
experience of the rigours of ordinary life in the
depressed 1930s or Word War II and whose
education in narrow neoclassical economics
(uncontaminated by the humanities or even by
the other social sciences) in elite universities is
claimed to make them natural sympathizers with
New Right ideas.
It may be too that such a new generation of
econocrats and accountocrats would be much
more prone to “bureau-shaping”, as analysed by
Patrick Dunleavy ( 1985, 199 1); that is, the
propensity of top public managers to aim for
high-status analytic work in collegial elite units
and to distance themselves from front-line
supervisory roles in favour of a “super-control”
position which offers more job satisfaction and
less tedious routine. Once bureaucrats adopt
such preferences, there is nothing against their
interests in enthusiastically cutting service
delivery budgets down the line, or in breaking
up and deprivileging the world of public service
delivery, so long as the power and status of
central agencies is retained or augmented.
A response to fi scal stress and poor
macroeconomi c pwformunce?
Another common way of interpreting the
demise of the old PPA model is to link it with
changing social conditions, akin to the “loss of
habitat” theory of dinosaur extinction. NPM is
often interpreted as a response to fiscal stress
and resistance to extra taxes. Underlying the measures of government size can conceal as
onset of such fiscal stress may be the changes much as they reveal (Peters & Heisler, 1983).
in income level and distribution, weakening the But if we look at four conventional measures of
“Tocqueville coalition” for government growth government size (that is, government employ-
in the electorate; that is, an electoral majority ment as a percentage of total employment,
of voters at below-average incomes who stand government expenditure as a percentage of
to benefit from increasing public spending finan- GDP, social security expenditure as a percen-
ced from income taxes. A move towards a more tage of GDP and tax revenue as a percentage
diamond-shaped income distribution pattern of GDP), it does emerge that the two most
lays the conditions for a new tax-conscious slimline governments within OECD (Japan and
winning electoral coalition, and NPM can be Turkey) placed a low degree of emphasis on
represented as the approach to public manage- NPM during the 1980s just as would be
ment which fits this new tax-consciousness in expected. But not all outsize governments in
marginal electorates; for example, by keeping the OECD placed high emphasis on NPM in the
overall public-sector pay bill down by means of 198Os, and the medium-sized governments
performance pay rather than by across-the-board also varied considerably in the emphasis which
pay rises in the traditional style. they laid on NPM. So if government size plays
However, if NPM is best explained as a a part in determining NPM emphasis, it is
response to fiscal stress and government over- probably a subsidiary one rather than the single
load, we might expect NPM to be most strongly determinant.
developed in those countries which score Similarly, the link between macroeconomic
highest on government spending and employ- performance and the degree of emphasis on
ment an&or have a history of relativeIy poor NPM seems to be far from clear-cut. NPM is
macroeconomic performance on the conven- often interpreted as a reflection of 1970s’
tional indices of GDP growth, public debt levels, economic chickens coming home to roost (i.e.
inflation and unemployment rates (such as the record of past or current economic perfor-
Greece or New Zealand). mance), and a return to hard-headed realism.
Indeed, NPM has frequently been interpreted But there appears to be no automatic relation-
(for example, by trade union critics) as little ship between the emphasis piaced by diITerent
more than a means of slimming down big OECD countries on NPM and their level of
government, and saving on resources in the performance on the four conventional macro-
public sector. If slimline public management is economic indicators in the post-oil shock era of
what counts in competition among industrial the 1970s. i.e. 1974-1979.
(or post-industrial) states for economic advan- Table ii gives some indicative data on this
tage, we would expect to see convergence, with point. It is true that some of the macroeconomic
the countries which are least slimline at the success stories of the 1970s are found in the
outset making the most dramatic strides in low NPM emphasis group, as might be expected.
adopting NPM doctrines (because those are the Countries like Japan and pre- 1990 Germany are
countries which would have the most to worry in this group. But not all of the high performers
about in terms of comparative advantage). of the 1970s are in that group, and nor are all
Equally, countries with small-sized public the economic “basket cases” of the 1970s (in
bureaucracy might have proportionately less to term3 of overall scores on GDP growth per head,
gain from putting greater stress on such CPI growth and unemployment) in the high
doctrines. NPM emphasis group. And even if we relate
Finding useful indicators for government size degrees of NPM emphasis to current economic
is no easier than arriving at robust measures of performance (on the same basis) over the
the nature of party political incumbency. Indeed, 198Os, the same puzzles arise, as can also be
it is commonty observed that coventional seen from Table 4. It .seems that macroeconomic
THE NEW PUBLI C MANAGEMENT 103
104
C. HOOD
TABLE 4. NPM emphasis and economic performance 1974-1979 and 1980-1988
Economic
performance
category
Econ era
Hi
Medium Hi
Medium and
Mixed
NPM emphasis 19BOs
Hi Medium lo
1974-1979 19Bo-1988 1974-1979 19Bo-1988 1974-1979 19Elo-1988
JaPm
Sweden Sweden France AUSt l k i BBD BBD
(4)
AUSt i Norway
Japan
(1)
Finland
Norway
N2 (4)
Canada Ireland Greece
AUS Aus
(4)
(4)
Spain
UK (4)
Turkey
(1) (3)
(4)
Medium Lo U.K.
Canada
Lo
IaY
Port.
N2 (1)
‘dY
Port.
(4)
Ireland
(4)
Switz.
Greece
(4)
Spain
(4)
Turkey
(4)
(4) (4)
Source: analysis based on OEW Historical Statistics.
Key to economic performance indicators:
Hi = all available indicators in hi category;
Lo = all avallable indicators ln lo category;
Medium hi = 25% or more of available indicators ln hi category and no indicators in lo ategory;
Medium lo = 25% or more of available indicators in lo category and no indicators ln hi ategory;
Medium and mixed = (a) all available indicators in medium category; (b) available indicators dlsuibuted across all three
categories (hi medium and lo).
Notes:
( 1) No indicator avallable for unemployment rates for this case for this period.
(2) No indicator available for GDP growth per head for this case for this period.
(3) No indicator available for CPl growth for this case for this period.
(4) No indicator available for government debt relative to GDP for this case for this period.
performance alone is not sufficient to explain
the rise of NPM.
As with Table 2, the placings are indicative
and “broad-brush”. NPM scorings are as des-
cribed on p. 101. Economic performance
scorings are based on OECD Historical Statistics
1974-1979 and 1980-1988. Four conventional
series were used: average unemployment
rates as a percentage of total labour force;
average rates of growth in real GDP per
capita; average growth rate in consumer price
index; and average public debt levels as a
percentage of GDP. For each of these indices,
the countries were divided into “hi”, “medium”
and “lo” scorers by the method used as before
(i.e. more than 1 S.D. away from the mean
counted as “hi” or “lo”, and the remainder
“medium”), except that for the CPI index two
outkrs (Turkey.and Iceland) were taken out, in
that both had scores more than twice the mean
and the effect of including them was to put
almost all countries in the “medium” category.
THE NEW PUBLIC MANAGEMENT 105
I nitial endowment
Popular wisdom notwithstanding, there seem
to be important cases which do not readily fit
standard explanations for why NPM developed
in the 1980s. Another possible explanation is
the baseline, or initial endowment from which
different administrative systems start.
Specificially it could be argued that for an
administrative system to move signticiantly
towards NPM, it must be set up at the outset in
such a way as to provide both motive and
opportunity (conventional elements of detec-
tive fiction) for incumbent politicians to want
to shift the administrative system sharply in that
direction.
We could argue that motive in this case might
be expected to consist mainly in the promise
or hope of resource saving from the adoption
of NPM measures, and could therefore be
expected to be proportionately higher in a
context of outsize government and/or acute
fiscal stress associated with poor macroeconomic
performance than in the context of slimline
government and/or strong macroeconomic per-
formance. Opportunity might be expected to
depend on the existence of some “Archimedean
point” from which would-be reforming politi-
cians can influence the public sector as a whole.
For instance, in a country like Switzerland,
where even the number of Ministries (seven) in
the federal government is set out in the constitu-
tion and has not changed for 150 years, the
opportunity for politicians or top officials to
reshape public administration is relatively slight,
because there is effectively no difference bet-
ween constitutional reform and administrative
reform. But in countries like the UK, where
there is no constitutional check to admini-
strative reform and politicians at the centre can
change the entire system, opportunity is much
greater. It would therefore seem that a crucial
variable for opportunity is the extent to which
there is an integrated public service controllable
from a single Point and without significant
jurisdictional breaks (for example, without
independent public bodies beyond the reach of
control by a single set of elected politicians, like
the German Bundesbank).
TABLE 5. Public management baseline styles and propensity
to shift NPM-wards: a tentative hypothesis
Stress on
integration Stress on collectivism in service provision
of public
StTViC~ Lo Hi
Hi
Lo
(1) (2)
“Japanese way’ “Swedish way”
Motive for switch: lo Motive for switch: hi
Opportunity: hi Opportunity: hi
(3) (4)
“American way” “German way”
Motive for switch: lo Motive for switch: hi
00Dommitv: lo Oowrtunitv: lo
Table 5 puts these two aspects of initial
endowment together to identify four polar
types: what is labelled the “Japanese way”,
where public service integration in this sense is
high but collectivism (in the sense of the relative
size of the public sector in spending and employ-
ment) is comparatively low; the “Swedish way”,
where both public service integration and
collectivism are high; the “American way”,
where both integration and collectivism are low;
and the “German way”, where collectivism is
high but integration is low. These labels are
used as convenient shorthand terms, and it
is not suggested that each of these countries
exactly corresponds to the stereotype in all
particulars.
On this basis, some countries would be much
more likely to move NPM-wards than others, at
least in the first round of public management
reforms. For the polar type labelled the “American
way”, there would on these assumptions be
neither motive nor opportunity to make a major
shift NPM-wards. No one would be in a position
to order the changes, and in any case the gains
would be expected to be less than in an outsize
government system. For the “Japanese way”,
there would be opportunity but again little
motive because the system is starting from
the “small government” box. And for the
“German way”, there would be motive, but no
opportunity.”
Only in the polar type labelled the “Swedish
106 C. HOOD
way” w oul d there be both motive and oppor-
tunity: motive, because outsize government
makes resource saving of key importance in
conditions of growing fiscal stress, and oppor-
tunity, because there are central points of
leverage over the entire public service. Hence it
might be argued that countries in the “Swedish
way” box (like the U.K., France, the Scandinavian
countries and possibly the Netherlands) would
be the type most prone to make rapid strides
towards the development of NPM in the
1980s.
A “variable diffusion” model of this kind can
help to explain what more generalized explana-
tions of NPM on their own cannot: namely, why
a number of key OECD countries under
governments of different political stripes shifted
to NPM in the 1980s while others moved
relatively little in that direction. All of the OECD
countries which seem to have placed the highest
stress on replacing PPA with NPM during the
1980s started from the “Swedish way” box in
Table 5 at the baseline, and some of the
countries which put medium-to-high stress on
NPM, such as France and Denmark, also started
as “Swedish way” cases in this sense. Of course,
such an explanation is not independent of
habitat-change accounts stressing the effect of
fiscal stress and tax-consciousness, but it does
explain why a similar habitat change might
produce different effects in different institutional
systems.
CONCLUSION
Compared to the voluminous literature on
deregulation and privatization, accounts of the
rise of NPM are less developed and sparser. We
still lack clear measuring rods for comparing
public management style, and the discussion
here is relatively speculative. Distinguishing
surface change from deep change will always be
difficult in public organizations. However, four
tentative conclusions can be drawn from this
highly exploratory exercise.
First, puce Aucoin, Osborne and Gaebler and
other “global change” interpretations, it is not
clear that the old PPA model of accountability
has collapsed everywhere, or to the same extent.
Even though we do not know how to measure
cross-national differences finely, there do appear
to be leaders and laggards in the process, and it
is interesting that some of the notable laggards
are leading countries in the international eco-
nomy, posing something of a challenge for the
view that public management intemationaliza-
tion parallels economic internationalization. If
a policy dinosaur is going into extinction here,
the process is still far from complete. Hence the
possible relevance of a variable difTusion model
based on initial institutional endowment, as
discussed in the last section. And if the different
institutional endowments lead to permanent
differences (rather than simply governing
the speed of PPA extinction), we may need
to be cautious about assuming that changes
in public sector accounting are likely to be
global.
Second, the conventional explanations of
change in the public sector do not on their own
seem fully to explain observed variations in the
degree to which NPM reforms were taken up by
OECD states in the 1980s. For instance, ideas
such as the view that NPM is all about right-
wingers in office, slimming down outsize govem-
ment or responding to macroeconomic failure
(in the past or present) all come up against
awkward cases which do not seem to fit
expectations when we look across the OECD
countries. But that does not of course rule out
such items as part of a broader multi-factor
explanation of the shift to NPM.
Third, there appears to be no simple relation-
ship between macroeconomic performance
levels and the degree of emphasis laid on NPM.
6 Indeed, Germany itself is an ambiguous case, since its “large government” character derives more from high public
spending than high public employment. If size of public employment is the key tn NPM motivation, there wnuld therefore
be no mare motivation than under the “American way”.
THE NEW PUBLIC MANAGEMENT 107
A working hypothesis might be that the show-
case economies were not under strong pres-
sures to shake up their public management
systems and that the more “basket-case” eco-
nomies lacked the capacity to do so, so that
those making the biggest strides with NPM are
likely to be those in the medium-to-poor
macroeconomic performance bands.
Fourth, there seems to be no simple relation-
ship between the political stripe of governments
(in so far as that can be gauged) and the degree
of emphasis laid on NPM. Are we to assume that
Downs’ (1957) classic ideas about policy con-
vergence in party competition better explain
apparent concensus on NPM than Hibbs’ ( 1977)
ideas about policy divergence? Or could it be
that apparently similar measures have been
adopted in different political circumstances for
diametrically different reasons and with quite
different effects? After all, such things often
happen in administrative reform. Perhaps the
classic historical case is merit hiring for civil
servants, which, according to Hans Mueller
(1984) was adopted by eighteenth-century
Prussia to bring the middle class into the public
bureaucracy and by nineteenth-century Britain
to keep them out.
On similar lines, it might be argued that NPM
has been adopted in some contexts to ward off
the New Right agenda for privatization and
bureaucide and in other countries as the first
step towards realizing that agenda. Much of NPM
is built on the idea (or ideology) of homeostatic
control; that is, the clarification of goals and
missions in advance, and then building the
accountability systems in relation to those
preset goals (cf. Dunsire, 1990). But if NPM has
itself been adopted for diametrically opposite
reasons in ditferent contexts, it may, ironically,
be another example of the common situation in
politics in which it is far easier to settle on
particular measures than on general or basic
objectives. That too may suggest that we should
be cautious about assuming that public-sector
accounting is likely to enter a new age of global
uniformity, at least in the sense of the wider
public management context within which it
operates.
BIBLIOGRAPHY
Aucoin, P., Administrative Reform in Public Management: Paradigms, Principles, Paradoxes and Pendulums,
Gouemunce(l99O)pp. 115-137.
Castles, F. G., Big Government in Weak States: The Paradox of State Size in the English-speaking Nations
of Advanced Capitalism, J ournal of Commonweal th and Comparati ve Politics ( 1989) pp. 267-293.
Castles, F. G. & Merri l l , V., Towards a General Model of Public Policy Outcomes, J ournal of Theoreti cal
Pol ftfcs ( 1989) pp. 177-2 12.
Corbett, D., Pol i ti cs and tbe Ai rl i nes (London: Allen and Unwin, 1965).
Douglas, J., Review Article: The Changing Tide - Some Recent Studies of Thatcherism. Brfti sb J ournal
of Pol i ti cal Scfence (1989) pp. 399-424.
Douglas, M., Cultural Bias, in Douglas, M., In theActi ve Voi ce, chapter 9, pp. 183254 (London: Routledge,
1982).
Dunleavy, P. J,, Bureaucrats, Budgets and the Growth of the State, Bri ti sh J ournal of Pol i ti cal Sci ence
(1985) pp. 29+328.
Dunleavy, P. J., Explaining the Privatization Boom: Public Choice versus Radical Approaches, Publ i c
Admfni stratfon ( 1986) pp. 1334.
Dunleavy, P. J.. Democr acy, Bumaucracy and Publ i c Choi ce (London: Harvester, 1991).
Dunsire, A, Con0vI in a Bureaucracy (Oxford: Martin Robertson, 1978).
Dunsire, A., Holistic Governance, Publ i c Pol i cy and Admi nfstmtfon ( 1990) pp. 3-18.
Dunsire, A. & Hood, C. C., Cutback Management i n Publ i c Bureaucraci es (Cambridge: Cambridge
University Press, 19B9).
Fox, A., Bqond Contract: Work, Power and Trust Rel ati ons (London: Faber and Faber, 1974).
Gorvin, I. (ed.), El ectfons Si nce 1945: A Worl dwi de Reference Compani on (Harlow: Longman, 19B9).
108 C. HOOD
Hall&an, J. & Wettenhall, R, Major Changes in the Structure of Government Institutions, in Power, J.
(ed.), PublicAdministration in Aust?nl&z A Watersbed, pp. 17-40 (Sydney: RAI PAAale and Iremonger,
1990).
Hegewisch, A., Public and Private Sector Trends in Remuneration Policies and Contract Fleslbility in
Europe, Paper presented at seminar International Come of Public SectorPay I n 10 Countries~
1990 Results j?vrn tbe Price Waterbouse0anfield Survey (London: Public Finance Foundation,
5 September 199 1).
Hibbs, D. A. Jr, Political Parties and Macroeconomic Policy, Am&curt Pollticul Science Reuiew (1977)
pp. 1467-1487.
Hirschman, A. O., Shifting I nuolwmen ts: Mu&e I nterest and Public Action (Oxford: Blackweli, 1982).
Hood, C. C., Public Administration, in Bogdanor, V. (ed), Fbe Blackwell Encyclopaedia of Poll&al
I nstitutions, pp. 504-507 (oxford: Blackwell, 1987).
Hood, C. C., Public Administration: lost an Empire, Not Yet Found a Role, in Leftwich, h (ed.), New
Ditvcttons in Polftical Scfence, pp. 107-125 (Aldershot: Elgar, 19901).
Hood, C. C., Beyond the Public Bureaucracy State? Public Administration in the 1990s (Inaugural lecture,
London School of Economics, 16 January 199Ob).
Hood, C. C., De-Sir-Humphrey-tying the Westminster Model of Governance, Go uemunce (199Oc) pp.
205-214.
Hood, C. C., A Public Management for Ail Seasons?, Public Administration (1991) pp. 3-19.
Hood, C. C. & Dunsire, A., Bureaumetrics (Famborough: Gower, 1981).
Hume, L. J., Bentham and Bureaucracy (Cambridge: Cambridge University Press, 1981).
Kamenka, E., Bureaucracy (Oxford: Blackwell, 1989).
Karl, B. D., fiecutiw Reorganfzatfon and Reform in tbe New Deal (Cambridge, MA: Harvard University
Press, 1963).
Marini, F., Towurd a New Public Administrution (Scranton, PA: Chandler, 1971).
Martin, J., A Profession of StatecraJ ?? Tbrtv J &says on Some Cumsnt hues in the New Zealand Public
Service (Wellington: Victoria University Press, 1988).
OECD, Public Management Developments (Paris: OECD, 1988).
OECD, Public Management Dewlopments (Pa& OECD, 1990).
Osborne. A. & Gaebler, T., Reinwnting Government: How tbe Enhqreneu~l Spfrft is Transf~ing
tbe Public Sector (Reading MA: Addison-Wesley, 1992).
Painter, M.. Values in the History of Public Administration, in Power, J. (ed.), Public Administrution I n
Australia: A Watersbed, pp. 75-93 (Sydney: RAIPAHale and Iremonger, 1990).
Peters, 8. G. & Heisler, M. 0.. Thinking About Public Sector Growth: Conceptual, Operational, Theoretical
and Policy Considerations, in Taylor, C. L (ed.), Wby Gowmmenfs Grow Measuring Public Sector
Size, pp. 177-197 (Beverly Hills: Sage, 1983).
Piore, M. J. & Sahel, C. F., The Second J ndxHrhal Divide (New York: Basic, 1984).
Pollitt, C., Managerialism and tbe Publk Services Tbe Anglo-American Ekperhce, 2nd Edn (Oxford:
Blackwell, 1993).
Power, M. & laughlin, R., Critical Theory and Accounting in Alveson, N. & Willmott, H. (eds), Critical
Management Studies, pp. 115135 (London: Sage, 1992).
Pusey, M.. Economic Rationalism in Canbewa (Cambridge: Cambrklge University Press, 1991).
Quirk, P., In Defense of the Politics of Ideas,jouml of Politfa (1988) pp. 31-41.
Schaipf, F. W.. The Political Calculus of Inllatlon and Unemployment in Weatem Europe: A Game-Theoretical
Interpretation, J ournal of Public Policy (1987) pp. 227-258.
Scott. G.. Bushnell, P. & Sallee, N.. Reform of the Core Public Sector: New Zealand Experience, Gouemunce
(1990) pp. 138-167.
Sjolund, M., Statens Liinepolitik 1966-19&l (Stockholm: Allmanna Fiirlaget, 1989).
Small, A. W., The Camerulfsts (Chicago: Chicago University Press, 1909).
Sturgess. G.. First Keynote Address, in Carey, G. & Ryan, P. (eds), In Tnansttiotz NSW and t&e
Corporatisatlon Agenda, pp. 4-10 (Sydney: Macquarie Public Sector Studies Progmm/Association for
Management Education and Research, 1989).
Taylor, J. A., Information Networking in Public Administration, Intern&o MI Review of Admlnistrat&e
Sciences (1992) pp. 375-389.
Taylor, J. A. & Williams, H., Public Administration and the Information Polity, Public Admdnfshation
(1991) pp. 171-190.
THE NEW PUBLlC MANAGEMENT
109
Thompson, M., Ellis, R & Wilds&y, A., Cultural Tbtwy (Boulder: Westview, 1990).
von Beyme, IL, Do Parties Matter? The Impact of Parties on the Key Decisions in the Political System,
Government and Op&nwition (1984) pp. 529.
Yeatman, A_, ‘Ihe Concept of Public Management and the Australian State in the 198Ck,Au.strulfun J ournul
of Publ i c Admi ni strati on (1987) pp. 339-353.
doc_390487559.pdf