[from ‘New Public Management, current trends and future prospects', eds. McLaughlin et al, Routledge, 2001,

isbn 0415243637, pages 57 - 76]



Chapter 4

Explaining the New Public Management

The importance of context


Norman Flynn









Ferlie et al. al. (1996) made a four-part classification of NPM approaches, based on four diagnoses of the problem by government. Their models were the efficiency drive; downsizing and decentralization; 'in search of excellence;' and 'public service orientation'. Peters (1996) also proposed a four‑part taxonomy of models of governing, market, participative, flexible and deregulated, which was also related to different diagnoses of the problems seen by governments. Our work (Flynn and Strehl 1996) of the same year shied away from trying to classify the management changes in seven European countries because we were still trying to identify the changes, their causes and consequences. We did, however, find some big differences among apparently similar countries and sought some tentative explanations for those differences. Ferlie et al. acknowledged different ,national type' contexts as well as sectoral differences while Peters's analysis, although geographically wide‑ranging, emphasizes the differences between sectors as the main determinant of the applicability of each model.


Much of the debate about convergence and difference has been concerned with sub-groups of the OECD countries, the English‑speaking countries sometimes referred to as Anglo-Saxon or Anglo-American (Kickert etc.) and Japan used as the only non‑Western example. A parallel debate has been carried on in the literature on public sector management and economic development. Kiggundu (1998) for example examines the context in which civil service reform in Africa has rarely been successful. Discussing the institutional capacity for good government in six poor countries, Hilderbrand and Grindle (1997) produced a typology of the economic, political and social conditions, which they call the (action environment', and the 'public sector institutional context' or the existing arrangements within the state.


McCourt and Minogue (2001) take an explicitly contingency approach to the explanation of differences between approaches to management. He posits a four‑part classification of approaches to public management, one of which is emerging and labelled 'strategic management'. The other three are 'Public Administration', the 'Washington Model' named after the 'Washington consensus' and NPM.


This chapter examines the proposition that different contexts generate different discourses, including different diagnoses of the problems thatgovernments are seeking solutions to. The diagnoses do not arise automatically from the 'objective' problems but are constructed within economic, political, institutional and cultural contexts. It also asks how the contexts shape the decisions that are made about what to do about the problems as defined. As with the diagnoses, it asks whether contexts have an influence on decisions which are not likely to be made on objective or 'technical' criteria. Some writers have argued that there has been a small set of decisions or solutions to the problems of managing the public sector and that these decisions are converging around a common set. Governments facing similar problems may arrive at similar solutions: after all, communication about what governments do is now easily available to policy makers and managers and there seems to be a convergence around a set of market‑oriented ideas about what governments do and how they behave and should behave. There has also been a counter‑argument that there persist differences among governments and what they do and how they do it. These differences are the result of contexts and are not simply the result of different stages of development towards a single, convergent ideal type, national conditions differ and therefore so do national practices. The arguments are not necessarily completely contradictory. At a very high level of generality, then all governments are similar: all (or almost all) are concerned with the efficiency with which their public servants work, the standards of public services, the deleterious effects of corruption on economic development. At a fine level of detail, differences will always persist. Researchers will always be able to find differences in budgeting processes or personnel practices or the degree to which competition is applied in service provision.


Both are correct and this chapter is not designed to contribute to either side of this perpetual dichotomy. It looks for explanations for the significant differences in how governments discuss their public sectors and behave towards them. It also asks whether there are elements of the context that make implementing change easier or harder. Its third ambition is to raise questions that might be useful in making international comparisons of public management and public management reform. There is a growing interest in comparisons and a growing number of territories used in the comparisons: Pollitt and Bouckaert (2000) include ten states in their study of public management reform. The purpose of comparative work varies. Our study tried to map the changes that were happening in management in seven European countries and tried to find patterns of difference and similarity. We tentatively offered some explanations for the differences in the constitutional and legal frameworks and in cultural attitudes to the state as constraints on management changes. Pollitt and Bouckaert link national politics to management arrangements and changes. Usually 'reforms' or significant changes in management arrangements are the 'dependent variable'. In this case what is to be explained is the diagnosis of the problems as expressed by governments and other agencies, the decisions made about what to do about them and the degree of difficulty experienced in implementation of the changes.


At a macro-level, the actions with which we are concerned include changing the portfolio of activities carried out by government and therefore the size of the public sector, measured either by the money spent or the numbers of people employed. The most evident of these is the privatization of state industries in manufacturing, public utilities, transport and telecommunications. These changes have been most apparent in Europe, although experience has been patchy: telecommunications has largely been privatized; railways are sometimes state owned, sometimes privately owned and sometimes in mixed ownership; many countries still have state-owned industries. Other states have privatized the railways, e.g. Japan, and many states have divested themselves of state-owned enterprises, especially if they have been through 'structural adjustment programmes' imposed by lenders. There has been little sign of privatization or a shift in the activities undertaken by the state in the United States or Canada.


At a micro-level, the changes with which we are concerned are the ways in which the remaining activities in the public sector are managed, including restructuring to create more accountable and manageable entities, the use of outsourcing, financial management and accounting, performance management, including customer service, and personnel management. In these areas there is a wide range of experience that has caused academics to doubt the notion of a single label of 'new'. The argument is that within general categories of 'reforming' or even 'NPM' there are differences between countries, in the three areas of discourse about problems and their solutions, decisions about what to do and in implementation. Differences in discourse may be less pronounced than differences in the other two areas, because there seems to be a fairly common 'meta-language' (Lynn 1998) that is used to describe, diagnose and prescribe management problems and solutions. The differences include:


§         the degree to which sectors or services are in practice transferred from the public to the private sector;

§         the degree to which markets are actually established, whether structures are decentralized in practice and the degree to which control is maintained in a small number of central institutions;

§         whether financial planning and control really switches from control over inputs to control through outputs or outcomes;

§         whether tiers of management are removed or merely re‑labelled;

§         whether the recruitment and promotion of staff is organized through a labour market or not;

§         whether customer orientation is based on empowerment of service users, on the development of direct democracy, on market‑research type consultation processes;

§         whether performance management is carried out through devolved decision-making and an emphasis on results or through centralized controls;

§         a variety of structural changes;

§         different degrees of reliance on NG0s in service delivery.


As well as differences within each of these areas, there are also differences in the emphasis that is placed on the elements in different governments and subnational government organizations. In some there is more emphasis on structural questions, in others on the mechanisms of financial accounting and management and in others on a management model based on empowerment of the front‑line implemented through 'cultural change' programmes.


This chapter tries to spread the net fairly wide in its attempt to catch or at least identify the independent variables in the contexts in which governments operate. Contexts are divided into external and internal. The definition of the external context extends slightly that of Hilderbrand and Grindle. It includes the macro‑ and micro‑economic environment; political conditions; political positions adopted by parties; the immediate problems facing governments and the institutional context in which changes are proposed, including the prevailing managerial climate, national culture, the socio‑technical systems in the sectors affected and the institutional capacity for change.


The purpose is not to define some formula for 'best practice'. The argument is that practice can be explained, to some extent, by reference to the context in which practices operate. It is also that the success of changes at the level of proposals and implementation is to an extent contingent. Certain practices are more likely to be successful than others, but that success is as likely to be contingent on the context as on the skill with which the practices are done. If these propositions are correct, then the choice of approaches to management should be informed by analysis of the contexts.


The argument is not that governments and others involved in the process currently follow the processes identified in the contextual model. The context affects the process but not necessarily at the level of a systematic analysis. Indeed. the process may well start with the choice of proposals, rather than an analysis of the problems to be solved. It has often been argued that proposals such as those implied in the 'Washington consensus' are a one-size-fits-all approach that pays little attention to the local context. The same can be true of locally developed solutions that may derive more from the demonstration effect of other governments' presentation of their successes in management reforms than from locally existing problems.



Macro‑economic conditions


Hood (1996), using four indicators of economic performance and a 'high, medium and low' classification of management changes, concluded that the level of economic performance was not a sufficient predictor of the level of implementation of NPM in the 1980s. Given the degree of variation in the adoption of the 'new' management methods, many other variables need to be included in the explanation, such as the size of government, the degree of integration of government, and so on. Hood finds that no single variable provides a good explanation for all cases and calls for research that includes a contingency and a diffusion approach. This has two implications for explanation. First, that there is a set of practices that can diffuse through different governments. Second, that as long as the contingent factors can be identified and measured, including how they change, they will explain changes in management practices. It could,on the other hand, be that sufficient explanation requires an understanding of the interaction of the variables, rather than treating each as an independent cause.


Macro-economic conditions do not automatically generate policy responses from governments. If there is an inflation target, price increases may trigger an interest rate rise, but the decision to control inflation through interest rates is the result of a political choice about the priority given to inflation over other economic indicators as a goal of government economic policy. For more complex political decisions as a result of macro-economic change, the political choices are more complex. For example, if there is an unpredicted fiscal deficit arising from reduced tax revenues in a recession there are various possible political responses. At one extreme, such as in Denmark in the late 1990s, the government can decide to meet all its previous spending commitments and fund them through a combination of tax rate increases and borrowing. At the other it can introduce austerity measures, reduce entitlements to state support and try to reduce the size of the public sector. Both are feasible political choices. In either case, there could simultaneously be a decision to introduce public sector reforms. In the first extreme case, it might be politically expedient to try to increase efficiency and reduce running cost expenditure as a gesture to those businesses and taxpayers who are asked to pay higher tax in a time of reduced incomes. In the latter case, saving on running costs might be part of the package of trying to reduce expenditure. There are many variants between the two extreme cases.


Take the opposite case: a government, such as Japan's between 1998 and 2001, that tries to take counter-cyclical macro-economic measures to reflate the economy. Since measures to encourage private spending failed, the government decided to increase public spending, especially on capital projects as a stimulus to the economy. Clearly, a period of fiscal loosening was required, despite the very large accumulated deficit. What effect might this have on the strength of efforts to reform the public sector? Little effort because of the fiscal loosening? Large effort because of the continuing recession and poor economic performance? In the event, the mediation of the political process, the loss of the hegemony of the LDP (Liberal Democratic Party) and the unpopularity of elements of the civil service all proved to be more important than the simple economic facts in explaining the calls for drastic cuts in civil service numbers and improvements in efficiency. The continued growth of employee numbers at local authority level since the start of the recession in 1989 also needs to be explained by local political factors rather than macro-economic management by central government. The deficit itself is of political origin as the LDP kept promises made at central and local level on spending and projects while unable to face the political consequences of raising taxes.


Our first category, shifting the boundary between the private and public sectors and changing the size of the state, is not a solution to short-term cyclical problems. It might be seen as a long‑term solution if it reduces governments' commitments to spend. Serious structural changes have never been achieved during the period of a cyclical downturn. If governments do decide to shift the boundary of the state for macro‑economic reasons it is because they have a view about what is appropriate for governments to do and how much resource they should control.


The appeal to the inevitability of 'globalization' and its impact on national economics and governments is such a case. New Zealand, a leader in both structural and management changes, made its public sector reforms as part of a general liberalization of the economy. The government claimed that an increase in productivity in the public sector would make a direct contribution to national competitiveness through its impact on the tax level. Corporatization of state activities and privatization of those for which there was no overwhelming case for them to be in the public sector were a major plank of the self‑imposed structural adjustment programme (see Kelsey 1995). The management changes within the public sector, including structural changes, introduction of markets, separating policy‑making from service provision, and so on, were also part of the package (see Boston et al. 1996). However, not all governments faced with problems of national competitiveness have taken this course of action. For example the French government used the need to improve national competitiveness as the reason for a series of 'modernization' efforts in the late 1980s and early 1990s, contemporaneous with the New Zealand changes but different from them.


Management changes can have little effect on budget deficits, especially in the short run. Reducing the running costs of public services, as opposed to transferring responsibilities to the private sector or to individuals, makes little difference to overall public spending except in those states where the level of spending is small in the first place and where most of the budget is spent on the salary bill. The major expenditures of European states, on welfare and education programmes, are unaffected by running‑cost‑productivity improvement.


In summary, when we consider the effects of the macro‑economic circumstances on public sector reforms, we have to distinguish between the short‑term efforts to reduce deficits and the long-term efforts to change the scale of state activities. We also need to consider the political processes that translate a budget deficit into government actions to cure it, or to wait until economic circumstances reduce the deficit by increasing tax revenues and reducing welfare expenditures. Different governments make different political choices. Governments faced with different macro‑economic environments may also reach similar conclusions, although for different reasons. The drive to reduce costs and increase efficiency was not confined to governments with budget deficits.



Micro-economic conditions


The micro-economic context also has an impact of what sort of changes are proposed and can be implemented. First, there is the question of the state of the supply side. If governments wish to increase outsourcing they have to have companies or third sector service providers to which to outsource. Efforts to generate competition for supply of goods and services where there is an established market, for example in building or engineering work, will be more successful than in those sectors where there is little supply, such as medical services in states where the government has an existing near monopoly control of hospitals, clinics and staff as is the case in most of Scandinavia. Examples of problems in generating suppliers include the attempts to generate private management of water supply services and fee collection where companies do not yet exist in this sector (see, for example, Batley and Larbi 1999).


The structure of the market also has an impact on the success of competition policy. Oligopolies in sectors such as large computer systems supply and weapons procurement reduce the cost advantages of outsourcing.


Second, there is the question of the labour market conditions facing governments that want to introduce more flexibility in their public sector labour markets. On the one hand, if there are few or no alternative job opportunities for displaced civil servants and other pubic sector workers, a programme of downsizing will meet with strong resistance. On the other hand, if there are labour shortages of skilled staff, a programme of liberalization of employment conditions will not improve the skill levels in the public sector. More generally, Hilderbrand and Grindle cite the overall level of development of labour skills as a constraint on the development of capacity for economic development.


The micro-economic environment is also dynamic. A market may be created if there is commitment to purchasing and if contracts are written and managed in such a way as to generate the supplying of materials to companies. On the other hand, mergers and acquisitions among supplying companies can turn competitive markets into oligopolies and cartels over time.



Political conditions


The organization of power relationships through political institutions also has an important impact on decisions and actions about the public sector. A commonly invoked condition for successful change in the public sector is the willingness of political leaderships to drive through radical changes. Governments with large parliamentary majorities, clarity about what they want to do and individual champions of change are said to be an important element in breaking down resistance and overcoming inertia. Governments that rely on coalition politics are more likely to compromise when making changes that affect the interests of coalition partners. Parties that rely on the support of people such as civil servants or professionals for their survival in government are also likely to be more timid about change. Examples include the Lubbers government in the Netherlands in the 1980s, which had a radical programme of reform but not a wide base of support in a system that requires the development of consensus before changes are made. Various attempts at decentralization of financial control in France have been countered by the powerful Ministry of Finance, whose power and influence was threatened by such developments. President Clinton's efforts to make radical reforms of the US health insurance system were defeated by an alliance of political opposition and the powerful insurance industry.


Attempts at civil service reform have also floundered on the rocks of political patronage and the capture of institutions by party factions. Reduction in the number of ministries and privatization of some functions in Japan was stopped because of the potential loss of power bases by important factions of the LDP. If political parties and their constituent factions are closely allied to specific institutions and their powers, it is more difficult to reform those institutions or to abolish or slim them. The Italian parliament has sufficient members of the academic profession to curtail managerial inroads into the universities.


Institutions or the professions that staff them may also have support from their own constituencies, other than through the political parties. If the medical professions are held in high esteem they can call upon public support in their resistance to unwanted managerial changes. The opposite may also be true. When teachers are held in low esteem, for example, education reforms are easier to implement than when they are cherished by other teachers and parents. The status of civil servants varies from country to country, from one extreme where they are poorly (and sometimes rarely) paid, to the other, such as in France or Singapore, where they are well paid and generally held in high esteem. Governments may manipulate public opinion by praising or vilifying public employees, but the process takes time. In some cases, popular support wanes because of the perceived performance of the institutions. The Ministry of Finance in Japan lost its high status gained from its perceived role in helping bring about fast growth when growth slowed and corruption scandals occurred.


Another protection against civil service reform is the constitutional position of the civil service in relation to the government. We found (Flynn and StrehI 1996) that where there were strong and detailed administrative and constitutional laws, governmental desires to bring about management changes were slower and had less chance of success than in states where ministerial actions could be implemented without legal and constitutional change.


The particular form that managerial change takes is also influenced by the relative power of particular ministries or commissions. When the Treasury is dominant (as in New Zealand in the early 1980s) reforms will have an accounting flavour. When personnel departments are important, emphasis is more likely to be on human resource policies and practices. It is interesting that it took over a decade of reforms in New Zealand before the personnel implications of the changes were confronted, regarding recruitment, retention and development of staff under the redesigned regime.


In some states trade unions are powerful and can have an influence on the form and implementation path of changes. Jamaica is one example.


These issues are about the nature and structure of national politics. They relate to some extent to what is called 'regime type', where types are defined as dictatorships, autocratic, democratic, and so on. However, it is not possible to read off the types of reforms directly from regime type. Even dictatorships rely on coalitions of forces to stay in power, and democracies have a variety of political processes. The general point is that the political structures and processes have an influence on the way that issues are raised and put on the reform agenda. They also mediate the processes by which solutions are found and policies are formulated and implemented. All change generates winners and losers, whether they are changes to the functions carried out by government, the size of the public sector, the way in which services are delivered or the management arrangements of individual departments and institutions. It would be unusual for there to be a complete separation between the people making the choices and implementing the changes and those affected by them. In some cases the people are indistinguishable, since they belong to the same families and went to the same schools. The upper reaches of the British, Japanese and French civil service all have close connections with the upper reaches of the legislatures and other branches of the state. This does not mean that change is impossible, but the nature and speed of the change will be affected by that political fact. Mutual obligations formed by the connections' will inform the diagnosis and influence the actions taken.


Where the connections are less close and there is clear demarcation between the parts of the state and between the state and the outside world, the political processes of change will be different but they will still take place.


Another important political condition is the degree to which people outside the country influence decision‑making. In the extreme, a debtor country has to submit to structural adjustment, privatization, management measures as well as macro‑economic policies in exchange for loans. There are also slightly more subtle pressures as small states operate to some degree as clients of the USA or ex-colonial powers, following policies that were formed outside the country in which they are being implemented. The mechanisms for this are technical assistance missions, the World Bank (and Inter-American Bank and Asian Development Bank), which import ready‑made solutions such as down‑sizing, outsourcing, privatisation, and so on, prior to diagnosis of the specific problems.


Political positions on the public sector


As well as the political structures and positions being important, the positions adopted by the parties about the public sector are also a crucial influence on the analysis, diagnosis and actions taken towards reforming the state. Obviously influential positions were those of what became known as the 'New Right', despite the fact that many of them were far from new. In brief these are:


§         that markets were more efficient than any other method of allocation and what could be left to the market should be;

§         that all motivations are selfish, so managers should not be allowed to make big decisions on budgets or services because they will only serve their own interests and not those of the people who they should serve;

§         that workers' motivations are the same and they should therefore be tightly controlled;

§         that self‑reliance is better than other‑reliance and therefore the state should do as little as possible.


The consequences of these positions for policy towards the public sector include privatization, cutting back state functions, dictatorial management and the search for mechanisms to bring managers under political control. For all the publicity that these basic positions received, they were adopted by very few countries and even in those where they were espoused or adopted, implementation programmes were limited to a few states.


Linked but not identical is the view that it is morally right for the public sector and its workers and managers to suffer hardship just as other sectors have suffered. Since industries are subject to intensified competition and are laying off staff, closing plants, restructuring, de-layering and asking everyone to work harder and smarter, then public sector workers should do the same. If politicians have close connections with business, they will be subjected to this moral argument and many will agree with it. The result of the position is that the public institutions should be subject to periods of staff cutting, restructuring, business process re‑engineering whatever the impact on productivity or quality.


There is also the doctrine of 'modernization', in which all the institutions, including businesses, democratic processes and the public sector, should contribute to some sort of national renewal and lead to greater national competitiveness in world markets. This position was taken by the ruling parties in New Zealand and France in the early 1980s, in Malaysia in the mid-1 990s and in Britain at the end of the 1990s. The consequence of this position is that there should be technical change and improvements at the customer interface level of the organization.


A variant is the view that the civil service and public sector in general are blocks to national and economic development. They probably cannot be turned into 'engines of change' (or some other mechanical metaphor) and should therefore be reformed sufficiently to stop them putting a brake on development. Corruption and/or unnecessary controls and 'red tape', which are part of the process of generating bribes, slow down economic development and building and should be rooted out. A variant is the US 'big government' line in which anything that government does is detrimental to entrepreneurship and should therefore be stopped. If this is not possible, then the public sector should become more entrepreneurial. The 'reinventing' campaign in the US was essentially legitimizing discretion at low levels of the organization in opposition to the rule‑based culture of 'big government'. It was about empowering the individual manager against the system, rather than trying to change the system.


An alternative is consensual politics and a corporatist approach. States with evenly balanced coalitions of a group of parties have to have consensus building processes or frequent changes of leadership and policy. The coalition building sometimes goes beyond parties and includes representatives of workers, business, churches and other sectors. The epitome of this approach is the Netherlands' consociational democracy' (Lijphart), which has also been identified in Malaysia (Jesudason (1996) coins 'syncretic state'), for example. In these conditions,radical change is unlikely, as no party is likely to produce a set of proposals outside the consensus. This is the case of reforms in both the Netherlands and Malaysia, which have consisted of changes in working practices, quality improvements, enhancement of the relationships between government and citizens, rather than radical programmes of privatization, restructuring or redefining the role of the state.' Consensus and corporatism concern political process but also have an influence on the positions likely to be adopted by the parties involved in the processes.


So far we have discussed the public positions of parties with regard to the public sector. Covert positions include individual, faction or party enhancement. When politics is organized mainly for personal gain, policies towards state institutions may also be designed to gain power and reward. For example, a privatization might be organized to benefit a faction, or it might be avoided if a faction gains more from the institution's position in the public sector.' Or a party may propose the creation of new institutions in opposition to those in which the party's position is weak. The distribution of powers and responsibilities among central and sub‑national governments is often organized to maximise ruling party control. When this form of power and reward‑seeking dominates politics, it is difficult to predict the implications for public sector structure and process reform which will depend on the relative positions of the players.



Immediate problems


Politics concentrates the mind on the short term. In democracies, elections are the time horizon within which decisions are made. In autocracies, public support is nurtured by visible events and displays. Policy is often made in response to immediate events, especially negative ones. Spectacular failures of the civil service or public authorities generate demands for change. The response of the authorities to the victims of the earthquake that struck Kobe, Japan, caused a huge drop in confidence in the government and the ruling party and set off a search for reforms. The economic crisis in Asia produced calls for structural reforms in the relationship between government and the private sector in many countries. The original reforms in New Zealand were triggered by a balance of payments crisis.


Other events can be less spectacular and of longer duration. Failures to produce adequate urban water supply, long delays in registration of births, prolonged theft of public funds, visible bribery beyond the norm, all put pressure on politicians to do something to improve the situation. Sometimes the pressure comes from citizens as individuals, sometimes as pressure groups and sometimes from businesses, either within or outside the government's networks. Very poor performance, or absolute failure, generates the necessity for some action. At a lower level failure may be less spectacular but still apparent. It becomes obvious when an education or health system is not working to the satisfaction of the service users. Consistent failure leads to legitimation problems for the government and/or the ruling party or parties.


Among immediate events we can also count fiscal crises: periods in whichthe revenues collected are insufficient to pay the bills. Although the causes may be cyclical, predictable and temporary or structural and permanent, the occasion of a deficit and consequent borrowings (or, in extreme, failure to pay salaries of state workers) triggers a policy response. If the fiscal crisis results in borrowing, then the urge for reform is reinforced.




All the above conditions vary from country to country. When governments decide what to do about the public sector, they are influenced, at least, by all of these factors. The diagnosis of the problem to be solved emerges as a response to the interpretation of the short- and long-term problems and is mediated by the political process.


Various ideal-type diagnoses have emerged. Peters (1996) identifies four: the monopoly position of government agencies in some services; the impenetrability of hierarchy; the inflexibility of permanence; the rigidity of excessive internal regulation. We could add to these: the state sector is too big because it is doing things that it should not be doing; public services are so unresponsive and ineffective as to generate public dissatisfaction with government; the public institutions, although performing the right functions, are inefficient and could do more with less; the public sector is hindering economic development and growth.


Not all these diagnoses are mutually exclusive and may appear in various combinations.


The diagnoses do not arise automatically from some 'objective' conditions. They are interpretations both of what is wrong and of what might be done about it. Take Peters's first: the diagnosis that governments should not have monopolies. In some states the question of a state monopoly of prisons, for example, does not arise, whereas in others private sector companies run most of the prisons. In those countries where the state runs all the prisons, monopoly as a diagnosis of the problem with prisons would only arise if there were the political possibility of alternative forms of provision. Or take the second: the fact that hierarchy prevents participation of workers and citizens in the decisions made by the organization. If hierarchy is a cultural norm and has been established for acceptable reasons, the existence of the hierarchy would not arise as the diagnosis of a problem.

  And so on through the list. The diagnoses of the problems arise from the context. The context shapes the diagnoses, the solutions proposed as well as the possibilities of implementation. The three elements are closely connected. Diagnosis, as we see in the Peters example, presupposes a type of solution: competition as a cure for monopoly; flat structures as a cure for hierarchy. Diagnosing the state as too big presupposes re-drawing the boundaries; inefficiency presupposes cutting budgets. In extreme, the cure may be in search of the problem. It has been said that lenders and consultants have pre-selected solutions to problems they have yet to find.  

Similarly with reform proposals and implementation. Successful implementation of reforms requires either that the change goes with the flow of the context, or that the relevant element of context can itself be changed to allow the reform to occur. Sometimes the proposals and implementation plans are so far out of line with the context that it may be questioned whether they were ever serious in the first place and were not simply a rhetorical device to satisfy some important stakeholder.


There are three elements of context that also influence the managerial solutions chosen and the success of their implementation. The first is what might be called the management climate or ideology. This means the approaches to management that are current, known to politicians and others designing reforms and likely to be acceptable to workers. The second, which has an impact on the first, is certain aspects of national culture that have an influence on the organizational cultures of the public sector. If management arrangements contradict the prevailing culture they are less likely to succeed. Third, there is a socio‑technical system that operates in the sector(s) concerned. As in business organizations, where management solutions that are appropriate in a nuclear power plant might not work in a hotel, so solutions that are appropriate in prisons might not work in tax collection departments.



The institutional context


The management climate or ideology


Management climate varies with geography, time and sector. It consists of what companies and other organizations do, what academics and consultants teach, advise and write about and how management is represented in the media through stories of success and failure and the creations of heroes and heroines and villains. Politicians concerned about their approach to management are faced with libraries full of books, consultants keen to be hired and a general climate about what is currently in favour. They have to make some choices.


Some of the choices have been clear for half a century or longer. The 'human relations' school versus the close supervision and piecework school is till a debate that is acted out in choices about quality processes, whether to have performance-related pay, how much autonomy professionals should be allowed, and so on. Looking around management changes in the public sector you can see examples of both solutions.  

The centralization-decentralization debate has gone in cycles in management thought. What companies do about how much autonomy divisions or units they should have vis-a-vis head office depends both on the preferences of managers and on the context in which companies find themselves. Public sector changes have periods of centralization and decentralization. A famous example is that of the self‑managing schools experiment in Victoria State, Australia, which spread to many other places. The experimental arrangements have now been reversed by the government in Victoria because of its impact on equality of opportunity for pupils.


A third continuing debate is about the choice between vertical integration and specialization. The advantages of developing a high level of skill in a particular activity are weighed against the advantage of being able to control the whole of the production process. At any time, examples of both solutions can be found in the private sector. Outsourcing is an example of specialization in what is sometimes called 'the core business', and private sector experience of this is evinced in support. Examples of vertical integration, such as the European holiday industry with its vertically integrating bookings, airlines and hotels, are ignored.


Approaches to quality also vary. There are two opposed positions: that quality is best achieved through detailed design and control; and that quality is achieved through empowerment and participation. Both are used in approaches to quality in the public sector, sometimes simultaneously.


A related issue is whether products and services should be standardized (Fordism) or individually designed for each customer, or at least having the appearance of being such (post-Fordism). Manufacturing technology offers certain choices of design and finish that allows customization: should services also be customized?'


Apart from these five fundamental choices, there are fashions that arise and are promoted from time to time. Business process re-engineering is one example of a management practice marketed and promoted by two authors, Hammer and Champney, as if it was a new, discrete management process. More recent is the fashion for 'strategic alliances', 5 which emphasizes cooperation over competition, as a source of success. The 'new economy' has also become fashionable, emphasizing use of the internet in service delivery, call centres and efficient anonymity as a source of customer satisfaction. Governments are busy smartening up their web pages and installing or contracting for call centres.


Politicians are exposed to these ideas and trends and filter them through their decision‑making processes. The language of the fashionable ideas quickly appear in documents, frequently with genuflection to 'private sector practice' as if there was a single set of good practices in business.



National cultures


Hofstede (1980, 1991) and Hampden-Turner and Trompenaars (1993) analysed large samples of managers to see if there was a pattern of national cultural characteristics that had an impact on the way organizations are managed. Hofstede's sample showed four consistent dimensions: attitudes to authority; the relationship between the individual and the group; 'masculinity' and 'femininity' attitudes to uncertainty. Hampden‑Turner and Trompenaars identified seven dilemmas: universalism versus particularism; analysis versus integration; individualism versus communitarianism; inner‑directed versus outer‑directed orientation; time as sequence versus time as synchronization; achieved status versus ascribed status; equality versus hierarchy. The two lists overlap somewhat.




In both cases, clusters of characteristics were assigned to countries. If management processes or management changes go against the flow of these clusters of characteristics they are less likely to be successful than if they go with the cultural flow. Although no detailed studies have been made of the impact of national cultures on public sector management, it is possible to surmise the likely impact of culture on specific management reforms.


Authority and power distance and achieved status and acquired



Some reforms have involved reducing hierarchies and making individuals responsible. Recruitment practices that open up labour markets replace promotion through seniority with appointment on merit and performance. The cultural limits on these changes are that the prevailing beliefs may be challenged. If staff are unwilling to accept authority from those for whom they have no respect or who do not have sufficient seniority then the old hierarchies will reestablish themselves. A related cultural feature is the willingness to accept responsibility for decisions. A hierarchy which passes memoranda up and down to spread responsibility is fundamentally challenged when authority and responsibility are forced on individuals.


The individual and the group and individualism and



One aspect of personal responsibility is linking pay and promotion to individual performance. Where the individuals subsume their individuality within a group identity, individual appraisals and individual performance pay schemes will become an empty ritual.'


Masculinity and femininity

These terms are shorthand for styles of working and managing. They cover dichotomies between competition and collaboration, caring values versus material success, and so on (Hofstede 1991: 79‑108). They do not imply that all women have feminine characteristics or that all men have the opposite.


Management changes in the public sector have involved both changes from 'feminine' to 'masculine' and vice versa. Regimenting the caring professions is an example of the former. Promoting collaboration rather than competition might represent the latter. Crudely, one would expect changes of the former sort to be effective in places that have high 'masculinity index scores' and the latter to be more successful in those that have low masculinity scores.


Universalism and particularism

This index relates to the preference for applying rules universally to all cases and making individual choices for particular cases. Clearly changes that increase individual discretion will have problems in cultures wedded to the universal application of rules.


Analysis versus integration

This dimension concerns the tendency to break problems into their component parts or to see them as a whole and find a whole solution in the broad context. A tendency to move towards management through effectiveness and management of outcomes clearly requires integration rather than analysis. A tendency towards analysis will favour management by outputs and individual units and activities.


Inner-directed versus outer‑directed orientation

Inner-directedness is defined as a preference for looking inwards in decisionmaking as opposed to taking signals from the environment. Obviously it will be easier to create and manage outward‑looking organizations in cultures that are conducive to that way of working.


Equality versus hierarchy

Creating teams of equals, project groups with no formal leader, quality circles and other techniques of the 'human relations' school will be easier in societies that are comfortable working in a non-hierarchical way.


It is not the intention to produce national culture stereotypes: Hampden-Turner and Trompenaars take the trouble to avoid such a charge and even distinguish cultures in different states in the USA. What such a set of categories does, however, is highlight dimensions of underlying beliefs that management changes might encounter. The reforms can either be designed to comply with the prevailing culture or engage in a culture change programme to alter it. Such programmes are not trivial since they have to deal with deep-seated beliefs.


The socio‑technical system


Ferlie et al. said that the sector in which management changes are implemented makes a difference to the success of the changes. As in the business sector, the variety of products, services, technologies and skills generate a variety of management approaches. Since the industrial revolution it has been obvious that production lines require/are based on the division of labour. Armies require hierarchy and obedience. Policing requires a degree of autonomy for the officer at the scene of a crime. Courts have to comply with laws and exercise their defined discretion. Tax collection requires equal treatment of taxpayers.


Often management changes are expected to be generalized across all public services and all sectors. The anomalies soon become apparent. A pathology laboratory worker famously responded to a 'customer care' initiative that she never saw more than a few dozen cells of any patient. Any changes need to take account of the nature of the service delivery system that is involved.


The main variables are: Is the service routine and rule-bound or does it require the operation of discretion in everyday decision-making? Does the task have work processes that come from the profession rather than the organization (medical, engineering etc.)? Does the technology determine the working arrangements? Are alternative technical delivery systems available? Is the work to uncontroversial performance measurement?


The answers to these questions will form a strong context in which decisions about increased decentralization, devolved decision-making and managerial discretion are made. They will also provide strong clues about how management control and accountability can best be organized. General principles of management, whether derived from the private or public sectors, do not provide specific answers about how services should be managed. In addition to the above contextual conditions, the content of the work is crucial.


Institutional capacity


A major constraint on management changes is the ability of the people to carry them out. Reforms that rely on new costing and accounting arrangements need accountants and book‑keepers to carry them out. Performance management requires skills from middle managers. Skills can be developed by training or acquired by hiring people, but both processes take time and may lag behind the reforms, making implementation slow and unsuccessful in the early years.


Crucially, a decision to provide services by outsourcing needs not only a range of suppliers but also the ability to write and monitor contracts with those suppliers. Batley and Larbi (1999) identified the lack of skill in this area as a constraint on the development of private urban water supply management in their sample of poor countries. The difficulty of being a 'smart' purchaser has been identified as a constraint on the development of contracting‑out and market solutions (see, for example, Walsh), especially when the services to be outsourced are complex and specialized.


Desired outcomes


The real purposes of management changes is perhaps the hardest item to disentangle from the language in use in official documents. The litany of 'flexible, customer‑centred, responsive' and so on is always a publicly stated aim. The politics of management change varies according to the political configuration: reorganizations can be used to benefit different factions; some reforms are designed to change the balance of power between politicians and professionals.


The ultimate outcomes are to gain or regain political support, whether from the electorate or business, or other interests. The support may be gained by cutting costs (and possibly taxes, although this has rarely happened) or improving services. More radical changes, such as moving the state‑private sector boundary, might well also be an ultimate objective, introducing more opportunities for business to operate in the areas previously occupied by the state.


In the real world of politics, the connections between desired outcomes, diagnosis and proposals maybe stretched or tenuous. Proposals may arise directly from the political process and the objectives may remain covert.


Reform proposals normally arise from some specific event or problem. A scandal exposes weakness; a deterioration in service reaches the point of intolerance by the public; an economic crisis exposes a country to outside influences; a party sees electoral advantage in improving in public services Whatever the event, the politicians and senior public servants, often advised or informed by colleagues in other countries, and other outside agents such as consultants, look for solutions to the problems. The desired solutions are at different levels, from changing the role and functions of the state to increasing efficiency and improving customer service. The outcomes from those solutions may also vary: they may be to gain electoral advantage for the ruling parts, to wrest control over public services from professionals to politicians; to cut taxes: to improve economic growth.




The argument has been, first, that there are differences among countries and sectors in the way management change is approached. Although at a highly abstract level, governments use similar language to describe their reforms, in practice there are different priorities and different objectives.


The context in which management changes take place account in part for the differences in the proposals and the success of their implementation. The elements of the context are summarized in Figure 4. 1. The importance of the elements varies among countries, and each country will have its own particular set of contextual elements. Explanation of a particular reform process and its implementation requires the identification of the relevant context.


Implementation of the solutions also takes place in a specific institutional context, defined here as having four parts: the management climate, the national culture as it impacts on organizational culture, the socio-technical systems in the sectors to be reformed and the institutional capacity for change.


The positive and negative elements of the context need to be identified if this analysis is to be used to improve the chance of successful management changes. Some negative elements will be easier to change than others. For example, if there is a strong national culture that reinforces hierarchy and is comfortable with large power differences and reluctant to individualize responsibility, then a reorganization that removes tiers of management and devolves responsibility will be difficult. Other approaches to performance improvement, such as a hierarchical system of measurement and a collective responsibility, would be easier to implement. Devolved management is much more likely to be successful in a cultural climate that emphasizes individualism, competition and the acceptance of individual responsibility.


The context is not only important for the likely success of implementation but also for the sorts of solutions that are proposed. Basic political positions about the role of the state and the market strongly colour the attitude to the big questions about state-private boundaries. Positions about motivation and attitudes to the workforce also colour the types of management arrangements that are put in place.


In addition to political positions or attitudes, there are underlying structural attributes of the political conditions, such as the degree of patronage, the relative strength of ministries and professions, that affect the process of managing and of managing changes.


All this might seem a heavy burden on people trying to bring about change. Inertia would seem to have the best chance of success. 'Path dependency' taken to extremes allows no options. That is not the purpose of the argument, rather that if we want to understand the processes involved in changing management in the public sector we need to take account of all the contexts. Practising politicians and managers know this, for their efforts are aided or thwarted by the contexts in which they work. The contexts help to explain not only the different management arrangements in place but also the different goals and problems in achieving them.


1 Where these connections are so strong that the boundary between the state and the society is invisible and there is a strong network of connections, the state might be referred to as a 'network state' (see Flynn 1999: Chapter 6).

2 Counter to the global trend, the Netherlands government has consolidated third sector pension schemes into a state sector, while other places are seeking non-state alternatives.

3 Japan's Postal Savings Bank is a case in point.

4 See 'Towards a Post-Fordist Welfare State?' for a discussion of this.

5 See, for example, Doz and Hamel (1998) and Child and Faulkner (1998). 6 There is evidence of this, e.g. among Chinese workers in Hong Kong.


Batley, R. and Larbi, G. (1999) The Role of Government in Adjusting Economics, Paper 41, University of Birmingham.

Boston, J., Martin, J., Pallot, J. and Walsh, R (1996) Public Management, The New Zealand Model, Oxford, Oxford University Press.

Burrows, R. and Loader, B. (eds) (1994) Towards a Post-Fordist Wellare State? London, Routledge.

Child,J. and Faulkner, D. (1998) Strategies of Co-operation: managing alliances, networks and joint ventures, Oxford, Oxford University Press.

Doz, Y.L. and Hamel, G. (1998) Alliance Advantage: the art of creating value through partnering, Boston, Harvard Business School Press.

Ferlie, E., Pettigrew, A., Ashburner, L. and Fitzgerald, L. (1996) The New Public Management in Action Oxford, Oxford University Press.

Flynn, N. (1999) Miracle to Meltdown in Asia: business, government and society, Oxford, Oxford University Press.

Flynn, N. and StrehI, E (1996) Public Sector Management in Europe, London, Prentice-Hall.

Hampden-Turner, C. and Trompenaars, F. (1993) The Seven Cultures of Capitalism, New York, Doubleday.

Hilderbrand, M.E. and Grindle, M.S. (1997) Building Sustainable Capacity in the Public Sector. In Grindle, M.S. (ed.) Getting Good Government, Boston, Harvard Institute for International Development.

Hofstede, G. (1980) Culture's Consequences, Beverly Hills, Sage.

Hofstede, G. (199 1) Cultures and Organizations, Maidenhead, McGraw Hill.

Hood, C. (1996) Exploring Variations in Public Management Reforms in the 1980s. In Bekke, H.A.G.M., Perry, J.L. and Toonen, T.A.J. (eds) Civil Service Systems in Comparative Perspective, Bloomington, University of Indiana Press.

Jesudason, J.V. (1996) Malaysia: the Syncretic State. In Rodan, G. (ed.) Political Oppositions in Industrialising Asia, London, Routledge.

Kelsey, J. (1995) Economic Fundamentalism, London, Pluto.

Kickert, W (1995) Public governance in the Netherlands: an alternative to Anglo-American Managerialism, Administration and Society, 28 (1).

Kiggundu, M. (1998) Civil Service Reforms: limping into the twenty-first century. In Minogue, M., Polidano, C., Hulme, D. (eds) Beyond the New Public Management, London, Edward Elgar.

Lijphart (1985) Consociational Democracy, mimeo.

Lynn, L. (1998) The New Public Management: how to transform a theme into a legacy, Public Administration Review, 58 (3): 231-237.

McCourt, W and Minogue, M. (eds) (200 1) The Internationalisation of Public Management: reinventing the Third World State, Cheltenham, Edward Elgar.

Peters, B.G. (1996) The Future of Governing: four emerging models, Lawrence, KS, Kansas University Press.

Pollitt, C. and Bouckaert, G. (2000) Public Management Reform, Oxford, Oxford University Press.

Walsh, K. (1995) Public Services and Market Mechanisms: competition, contracting, and the New Public Management, Basingstoke, Macmillan.

back to shorter pieces      back to Norman Flynn’s homepage