For Mary, Ben and John PUBLIC POLICY An introduction to the theory and practice of policy analysis WAYNE PARSONS Professor of Public Policy and Head of the Department of Politics Queen Mary, University of London, UK Edward Elgar Cheltenham, UK • Northampton, MA, USA 2 PUBLIC POLICY sophicaľ frameworks which have had an influence on the normative, ethical, and methodological approaches to the theory and practice of public policy. A major concern in the study of policy is the actual 'process' of policy-making. In section 1.7 we consider some of the main approaches to the study of policy-making and policy analysis. At this point we need to address wider methodological issues. Having used terms such as 'frameworks', 'metaphors' and 'models', we must consider what these concepts mean and how change and 'shifts' occur between different frameworks or 'paradigms' (1.8). The penultimate section (1.9) applies the idea of shifts in paradigm shift to account for the changing focus of policy analysis in the period from the 1950s to the present. The final section (1.10) considers how the policy process has generally been 'mapped' into stages and cycles, and outlines the approach taken in the remainder of the book. 1.2 'Public' and 'policy' as concepts The starting point for a discussion of public policy must be to consider what we mean by the idea of 'public', and to account for the development of the concept in theory and practice. This is particularly important in view of the fact that the idea of the 'public' has undergone considerable change in recent years in the Anglo-Saxon world as elsewhere. ♦ Key texts Three useful texts for the analysis of the idea of 'public' are: S.I. Benn and G.F. Gaus (eds), Public and Private in Social Life, 1983: on the ideas of public and private. J. Habermas, The Structural Transformation of the Public Sphere, 1989: on the development of the public sphere. J.A.W. Gunn, 'Public Opinion', 'Public Interest', 1989: on 'public opinion' and 'public interest'. ♦ Let us begin with some terms in common use: • public interest • public opinion • public goods • public law • public sector • public health META ANALYSIS 3 • public transport • public education • public service broadcasting • public accountability • public toilets • public order • public debt We could argue that 'public policy' has to do with those spheres which are so designated as 'public', as opposed to a similar list we could make of expressions which involve the idea of 'private'. The idea of public policy presupposes that there is a sphere or domain of life which is not private or purely individual, but held in common. The public comprises that dimension of human activity which is regarded as requiring governmental or social regulation or intervention, or at least common action. Does this sphere of the public require a different form of analysis to that of the private, or of the business world? What is the relationship of the public to the private? What should be public and what should be private? To any student of modern politics these questions are all too familiar. However, the relationship of the 'public' and the 'private' is an enduring theme which we may trace back to the beginnings of civilization. In this section we shall briefly outline some of the main features of the development of the concepts in Western society and endeavour to show how a knowledge of the history of the ideas provides an essential background for the student of public policy in the late twentieth century. As we shall see, there has always been a tension or conflict between what is held to be 'public' and what 'private', and it is vital that in studying 'public' policy we set our present arguments in this wider historical context. A suitable starting point is that of ancient Greece and Rome. It is from the Romans that we derive our concept of public and private: they defined the two realms in terms of res publica and res priva. The Greek idea of public and private may be expressed in the terms Koinion (roughly, public) and Idion (equally roughly, private). Hannah Arendt's analysis of the Greek dichotomy of public and private may be summarized in the following set of opposites (cited in Saxonhouse, 1983: 380): public private polis household freedom necessity male female equality inequality 4 PUBLIC POLICY • immortality mortality • open closed However, as Saxonhouse notes, this is a somewhat over-simplified characterization of the boundaries or lines of demarcation between the two realms. In theory and practice the relationship was more complex and reflected the 'tragic' interdependence of the two spheres. Indeed, Saxonhouse argues that there was no unified conception of the relationship between the two, and a study of the literature of ancient Greece suggests that at least seven quite distinct conceptualizations of the tension between the conflicting demands of the public and the private may be discerned. It is in the work of Aristotle that we find the earliest attempt to find some kind of resolution to the conflict between the public and private in the idea of the 'polis' as the highest form of human association. This search for some arrangement whereby the tension between the public and the private may be resolved or mediated was to echo down the history of political thought to the present day. In the nineteenth century this resolution between the public and private spheres found its most powerful formulation in the ideas of the political economists. It is this formulation of the 'problem' of the relationship of the public and private spheres which continues to predominate in contemporary arguments about the role of 'public' policy. For the political economists the trick of resolving the tension as between public and private in terms of 'interests' was in the deployment of their idea of markets. As Habermas argues in the early nineteenth century, the 'public sphere' developed in Britain out of a very clear demarcation between public power and the realm of the 'private' (Habermas, 1989). Through market forces the maximization of individual interest could best promote the 'public interest'. The free functioning of individual choice and freedom could, it was argued, advance both the interests of individuals, and the public good and welfare. The role of the state and politics was thus to create the conditions in which the public interest could be so secured. Government was consequently best when it did the least. For the political economists this did not mean that the state should not be involved in the provision of 'public' facilities, but that the crucial line of demarcation was economic freedom. The public interest in this sense was most likely to be served when the interests of economic freedom and the market were facilitated by the state, rather than being constrained or regulated. Order was essentially a spontaneous outcome of private choices. Public intervention was primarily desired so as to secure a framework of law, rights and order, rather than to interfere with the natural equilibrium which was the outcome of self-interest. Private interests were convergent with the public interest. META ANALYSIS 5 The quintessential statements of this view may, of course, be found in the writings of Adam Smith's Wealth of Nations (1776) and James Mill, Torrens, McCulloch and their popularizers (Parsons, 1989). This notion of the public as essentially a space which did not involve the interference in economic and business activities, and in which there was a well-defined boundary between the public and private spheres was, as Habermas shows, in contrast with the continental European tradition of the public as encompassing business and trade and 'private' life to a far greater extent than that which developed in Britain and America. In France and Germany, for example, the relationship between the state and business and trade was to be markedly different to that of the US and Britain. However, the liberal idea(l) of a clear distinction between the public and private began to collapse from the late nineteenth century onwards. The penetration of the public policy into what the political economists would have regarded as private took place in almost all areas of 'social life'. Education, health, welfare, housing, urban planning were all to become subjected to regulation and /or state interference (see Heidenheimer et al., 1990: passim). This process of collectivization in the public domain took place at different times in various industrial nations, but always for tlie reason that certain kinds of problems were no longer seen as purely 'private'. J.S. Mill had, in the mid-nineteenth century, provided the essential criterion for this shift of boundary: harm. The private was that sphere which did no harm to others (Mill, 1968). The problem was, of course, that the notion of what counted as 'harm' changed and expanded as more and more information about social and other problems legitimated concerns about the public consequence of private actions and supported arguments for reform. The utilitarianism of Mill and Bentham also provided another important test for determining public policy: the greatest happiness of the greatest number (see below: 1.6). By the early twentieth century, the liberal conception of the 'public' and the 'private' was undergoing a profound change. The 'new liberalism' as expressed by Dewey in America, and Hobhouse and Keynes in Britain took issue with the idea that the market could any longer bring about a convergence in 'public' and 'private' interests, or left to itself could promote a spontaneous order. For both Dewey and Keynes it was knowledge - organized intelligence, as Dewey termed it - which could now provide the means by which private and public spheres and interests could be balanced and advanced: laissez faire had had its day. A more knowledgeable form of governance was, from the standpoint of this new liberalism, the key to resolving the conflict between the claims of the private and the public. This was, of course, no new idea: Plato had long ago also come to the conclusion that philosophers would make the best of kings. 6 PUBLIC POLICY It was in the context of this 'new liberalism' as articulated by Dewey and Keynes (and others), and expressed in the practice of Roosevelt's New Deal and war-time administration and reform, that the 'public policy' approach was to develop. As Lasswell notes, the policy sciences as developed after the Second World War were but an 'adaptation of the general approach to public policy that was recommended by Dewey and his colleagues' (Lasswell, 1971: xiii-xiv). In the postwar era liberal ideas about the purpose of public policy-making were predicated on the belief that the role of the state was to manage the 'public' and its problems so as to deal with those aspects of social and economic life which markets were no longer capable of solving. The key to this brave new world was the development of a policy process and decision-making which was more informed by knowledge than it had been in the past. The 'old' liberalism was by no means dead, but it showed little signs of life until the 1970s. The claim that knowledgeable governance could better 'solve' or mediate the relationship between public and private interests began to sour in the era of stagflation. 'Keynesian'-inspired economic management and liberal welfare reforms seemed to be creating more problems than they solved. At this point Adam Smith's invisible hand clenched its fist and struck back. The champions of the 'new [sic] right' were Hayek and Friedman. They argued, to great effect, that this attempt to use public policy to promote the 'public interest' was flawed. For, as the political economists of the nineteenth century had showed, the public interest could only be advanced through allowing private interests a free hand. The answer they and others in the 'new right' camp argued, was to contract the 'public sector' and expand the use of the market mechanism to better ensure that, where there was a public sector, it functioned in a way that corresponded to market, or 'private sector', principles of 'management'. One of the consequences of the growth of the state as a means of reconciling public and private interests was the development of 'bureaucracy' as a more rational form of organization (see Weber, 1991: 196-252). 'Public administration' evolved as a means by which the 'public interest' could be secured through a neutral class of civil servants whose task it was to carry out the will of those elected by the people. Public bureaucracy was, therefore, different to that which existed in the private sector (business, commerce and industry) because it was motivated to secure the 'national interest', rather than private interests. Thus, whereas for the political economists (and the new right) only markets could balance private and public interests, the 'new liberalism' was based upon a belief that public administration was a more rational means of promoting the public interest. META ANALYSIS 7 It was in the 1880s, when Woodrow Wilson formulated the essential theory for this conceptualization of bureaucracy as a defender of the 'public interest', that he posited that there was an important distinction to be made between politics and administration (Wilson, 1887). Public administration as a framework for the analysis of bureaucracy in liberal democratic political systems thrived in this period when the civil servant was viewed as a functionary involved in the rational pursuit of public interests as defined by the political process. The idea of a rational, hierarchically arranged non-political form of administration was central to tlie idea of liberal democracy (see, for example, Mill, 1968) The division of the state into a political realm and a 'rational' or bureaucratic realm paralleled the demarcation between the public and private spheres. As the division of public and private began to to appear more and more imperceptible, the state laid claim to a legitimacy based upon its capacity to ameliorate a growing range of problems defined as 'public'. It was in this period - roughly 1950s-1970s - that public policy really began to take off, and public administration began to move into a state of decline which was to accelerate in the 1980s. Central to this change in orientation was the notion of rationality. Weber had shown that the growth of bureaucracy was due to the process of 'rationalization' in industrial society. The bureaucrat was the rational functionary who served the public interest. The rational public interest argument began to erode from the late 1940s onwards. It took three main directions: • Studies which posited that bureaucratic rationality (as set out by Weber) was a theory which needed re-examination. In both theory and practice it was demonstrated that bureaucracies exhibited a large measure of 'irrationality', or at least 'bounded' rationality (see Simon, 1945; Lindblom, 1959). • Studies which argued that in reality bureaucrats did not function in the 'public interest', but displayed the capacity to have distinct goals of their own (see Mueller, 1989). • Research which questioned the distinction between policy and administration (see Appleby, 1949). By the late 1970s the lines of demarcation between public and private and policy and administration were looking increasingly less well-defined. Thus as a leading comparative study of bureaucracy was to observe: "The last quarter of this century is witnessing the virtual disappearance of the Weberian distinction between the roles of the politician and the bureaucrat, producing what may be labelled a "pure hybrid"' (Aberbach et al., 1981: 16). The 'new' liberal 'solution' to 8 PUBLIC POLICY managing the relationship between the public and private interests through the state was looking less defensible in the light of the 'failures' of 'public policy' in so many areas. Two ideas of 'public' were questioned: the 'public' interest motivation of bureaucrats and professionals in the public service; and the relationship between the public (qua state) and private (qua market) spheres. It was undoubtedly the 'new right' (or 'old' liberals) who shaped much of the agenda for this debate. Returning to the position as set out by Adam Smith and the political economists of the nineteenth century, Hayek, Friedman and others asserted that the relationship between the public and the private was something which was best defined through the market and freedom of choice rather than by the state operating in the 'public interest'. During the 1980s and 1990s this argument that the demarcation between the public and private spheres should be left to the market has formed the dominant framework within which the theory and practice of public policy has taken place. This shift from the 'new' liberalism to the 'new right' in public policy may be discerned most clearly in the rise of the 'public sector management' approach and the demise of 'public administra tion'. ♦ The new liberalism and the public and private realms Several key texts provide the background to the new approach to the role of the state in seeing to public problems and interests: J.M. Keynes, 'The End of Laissez-faire', in Essays in Persuasion, 1926; and The General Theory of Employment, 1936. J. Dewey, The Public and its Problems, 1927. W.H. Beveridge, Full Employment in a Free Society, 1944. Keynes's essay on the end of laissez-faire is, perhaps, the most concise expression of the view that the state - armed with new knowledge and 'wisdom' - should seek to have a more interventionist role in social and economic problems. In the essay he surveys the development of liberalism from the seventeenth century onwards and argues that it was the economists who provided the 'scientific pretext by which the practical man could solve the contradiction between egoism and socialism which emerged out of the philosophy of the eighteenth century' (Keynes, 1926: 277). In the twentieth century Keynes thought that economists would have the leading role again, but that this time their theories would point towards a new kind of balance of interests. Progress, he believed, now lay in the path towards recognizing semi-autonomous bodies within the state 'whose criterion of action within their own field is solely the public good as they understand it, and from whose deliberations motives of private advantage are excluded' (Keynes, 1926: 288). The chief task, he argued, was to distinguish between those services which are 'technically social' from those that are 'technically individual'. Government should not METAANALYSIS 9 do those things which individuals can do already, but address those aspects of society and economy which cannot be 'technically' done by individuals. The seed of modern policy analysis is to be found in the notion that society should aim to improve the 'techniques' of governing a capitalist system, so as to make it 'more efficient' through 'wise management'. Whereas Dewey (1927) was to argue that experiments could nrovide the way to discover such new techniques, Keynes believed that it should come through 'thought', or the 'elucidation of our feelings', a 'candid examination of our own inner feelings in relation to the outside facts' (Keynes, 1926: 294). On this point Keynes's idea of 'policy analysis' is closer to Lasswell and Vickers than to the tradition of experimentalism and scientism, which may be traced back to Dewey. ♦ * What are the differences between the public and private sectors? Are the lines of demarcation between the two spheres as well marked as once they were? W.F. Baber (quoted in Massey, 1993: 15) argues that the public sector has ten key differences from the private sector: it faces more complex and ambiguous tasks; it has more problems in implementing its decisions; it employs more people with a wider range of motivations; it is more concerned with securing opportunities or capacities; it is more concerned with compensating for market failure; it engages in activities with greater symbolic significance; it is held to stricter standards of commitment and legality; it has a greater opportunity to respond to issues of fairness; it must operate or appear to operate in the public interest; it must maintain minimal levels of public support above that required in private industry. This focus on the 'profit' characteristics of the public sector and the 'non-profit' sector such as schools, universities, voluntary organizations, hospitals, etc., suggests to Anthony and Herzlinger that the line of demarcation is that: 'In nonprofit organizations, decisions made by management are intended to result in providing the best possible services with the available resources; and success is measured primarily by how much service the organizations provide and by how well the services are rendered' (Anthony and Herzlinger, 1980: 31). The 'non-profit' sectors thus are measured more by social welfare criteria than by financial profits. The authors argue that the non-profit sector may be characterized by: ' the absence of a profit measure; the tendency to be service organizations; the greater constraint in the goals and strategies they can develop; their greater dependence on clients for financial resources; their greater domination by professionals; their accountability, which is different to a private/profit organization; 10 PUBLIC POLICY • top management not having the same responsibilities or the same financial rewards; • the accountability of public-sector organizations to electorates and the political process; • their lack of a tradition of management control. However, as we shall see in section 1.5 below and in Part Four, the distinctions between public and private and profit and non-profit organizations is something which has undergone considerable change, due to reforms made within the public/non-profit sectors to make them more like the private/profit sectors. Against this view of a gradual convergence between sectors is the reminder offered by the Cabinet Office (1988, para. 1.5): Comparison with the private sector has to be treated with caution. In the private sector there is a direct relationship between commercial success ... and the standard of customer service. The public sector position is more complicated and in many instances distinctly different. In general, the reasons for providing a service in the first place, the nature of that service and the manner in which it is delivered, are not dictated by markets. In these circumstances the balance between public expectations and the level of service to be provided is decided on the basis of political judgements about economic and social priorities. All that said, those who execute public service functions have a professional responsibility to do so to the highest standards of service possible, within the given level of resources, and this is what civil servants want to achieve. ♦ ♦ But what is the public sector, exactly? One of the main frameworks for considering this question is provided by economists, who argue that we can analyse the public and the private in terms of 'goods'. A public good is a 'good' or service which is available to all. Pure public goods are those which are produced by the state, rather than by the market. Pure private goods are those which are consumed by choice and only those who pay for them may consume them. Samuelson (1954) suggests that the main characteristic of public goods is that they are indivisible, namely that they are available for all, and that they are non-excludable, unlike private goods which are, by definition exclusive. Public goods are paid for by taxes and borrowing, and their price may be expressed in the level of taxation required to finance their production. Private goods are paid for through a price system operated in a market. In broad terms it is possible to say that public choices do indeed involve decisions about 'public' or 'private goods'. In some countries, for example, health policy takes place in a largely public domain, where care is available to all citizens; and in others it is dominated by private care and personal health insurance. In some countries public transport is heavily subsidized; in others it is the case that public transport is practically non-existent, or the user has to pay a high price for his ticket. So up to a point the public/private good dichotomy has some utility in tackling the question of what is the public sector. However, what counts as a public good has long been (since the 1960s) a matter of considerable dispute amongst economists. Buchanan (1968), for example, questioned the purity of public goods as set out by Samuelson in 1954, and sug- META ANALYSIS 11 sted that there were many goods that could not be so neatly pigeon-holed. Some public geods c0U|d have 'excludable' benefits. Buchanan suggested that 'clubs' may exist which elude members of the public through a mechanism such as a toll or charge. Club theory merqed as an important aspect of the public-private debate, which pointed towards the fact ľhat in both theory and practice, the 'pure' public good was subject to (growing) impurity. From the point of view of the purity of public goods, public policy is really about defining what counts s public, who provides, who pays, how they pay, and who they pay. It does not follow that, because we admit that a service is 'public', it should be provided by the state, or that it should be open to all. In socialist China, for example, comrades pay for a variety of health and educational services. A public good may be privately provided, and consumed after a charge -or user fee - has been paid. A tank, for example may be ordered by the army, but manufactured by a private company. Furthermore a 'public' good may conform to the kind of criteria set out by Samuelson, but it may be available to people depending on criteria laid out in a policy: benefits which can only be distributed to defined groups or types of people. So-called 'merit' goods will exclude parts of the public on the grounds that they do not meet a set qualification or condition (Musgrave, 1959). In other words, the public and private sectors, when considered from the point of view of a theory of goods, reveal themselves as overlapping and interacting, rather than as well-defined categories. The public sector is a mix of public and private and of public goods which are rationed through a toll or by a criterion of merit. What determines this mix of public and private goods? Frey (1978) argues that there is a cycle in the demand for public goods which means that the public/private sector will over time change in response to the interaction of voters, government, civil service and producers (see Figure 1.1). Figure 1.1 The cycle in the demand for public and private goods Rise in disequilibrium Dissatisfaction with existing supply of public/private goods ■ Supply of goods Articulation of demand A new mix of goods Demands for a new distribution Reaction to demand Government reacts to demands Source: Adapted from Frey (1978:116-21) At some periods there were will be dissatisfaction with the way in which a public good is supplied or with its price. Government responds by changing the way in which a public good is supplied or the size or scale of its provision. This new mix provides the source of later issatisfaction. Is this what happened in the 1980s: a demand (from voters/capitalists for more 12 PUBLIC POLICY 'private' goods as against 'public' goods, a new supply (more 'club' goods, a reduction in 'pure' public goods)? Will the 1990s be about redefining the public/private mix? ♦ ♦ Whose public? Have men defined the 'public' sphere? Since the 1970s and 1980s feminist critiques of public policy have argued that policy-making has largely been framed by what men regard as the public domain. Carole Pateman, 'Feminist Critiques of the Public/Private Dichotomy', 1983 The dichotomy between the private and public is central to almost two centuries of feminist writing and political struggle; it is, ultimately, what the feminist movement is all about ... Liberal feminism has radical implications, not least in challenging the separation and opposition between the private and public spheres that is fundamental to liberal theory and practice. The liberal contrast between private and public is more than a distinction between two kinds of social activities. The public sphere, and the principles that govern it, are seen as separate from, or independent of, the relationships of the private sphere ... Feminists argue that liberalism is structured by patriarchal as well as class relations, and that the dichotomy between the private and public obscures the subject of women to men within an apparently universal, egalitarian and individualist order. (pp. 281-3) Kristie Beuret, 'Women and Transport', in Mavis Maclean and Dulcie Groves (eds), Women's Issues in Social Policy, 1991 Using data on travel methods, the age and sex of travellers, car ownership and use, the author argues that the lack of good public transport is a considerable disadvantage to the employment opportunities of women, and a factor which greatly limits their family and leisure activities. She concludes that women need to shape public transport agendas and policies if they are to improve economic and social opportunities. Beuret advocates a range of policies to this end, including: policies which improve the safety of public transport; better access to cars; travel schemes which make public transport cheaper and better geared to the needs of women; and more radical strategies to reduce the necessity of women to travel. The authors in Maclean and Groves's book point out that the boundaries of the public sphere have been constructed in such a way that the private problems which women as a whole face in transport and in other areas of social and economic life have been excluded from consideration as a 'public' issue', with the consequence that those areas which affect women more than men are poorly resourced and have low visibility and a low profile in the public sphere. Jan Pahl (ed.), Private Violence and Public Policy, 1985 This book examines the way in which 'public' policy in western societies has excluded the family from the public sphere. The result has been that violence against women in domestic circumstances has been a neglected area of policy-making. The issue illustrates the problem of defining public and private domains, and the relationship between public policy and individual META ANALYSIS 13 privacy. Pahl concludes that the way in which the private is defined (by men) tells us a great deal about power and powerlessness in society as whole. E. Meehan and S. Svenhuijsen (eds), Equality Politics and Gender, 1991 This is an excellent selection of papers dealing with the politics and policy of gender equality in a number of European countries. Several contributors discuss the impact that the public/private sphere dichotomy has had on the issue. ♦ The idea of 'policy' Words change their meaning. Like the notion of 'public', the idea of policy is, as Heclo (1972) argues, not a precise or self-evident term: To suggest in academic circles that there is general agreement on anything is to don a crimson in the bullpen, but policy is one term on which there seems to be a certain amount of definitional agreement. As commonly used, the term policy is usually considered to apply to something 'bigger' than particular decisions, but 'smaller' than general social movements. Thus, policy in terms of level of analysis, is a concept placed roughly in the middle range. A second and essential element in most writers' use of the term is purposiveness of some kind. [Heclo, 1972: 84) Despite this agreement, however, he notes that there are differences about whether policy is more that an 'intended' course of action. A policy may also be something which is not intended, but is none the less carried out in the practice of implementation or administration. In some languages, such as English, such distinction between 'policy' and 'administration' is well defined, in others it is not. The Oxford English Dictionary offers definitions of policy, covering: 'Political sagacity; statecraft; prudent conduct; craftiness; course of action adopted by government, party, etc.'. One dictionary of synonyms and antonyms offers the following: 'Policy, statesmanship, administration, wisdom, plan, role, action, tactics, strategy, sagacity'. And its antonym? Aimlessness. Dror notes that the notion of 'policy-making' as a 'conscious awareness of choice between two main alternatives for steering societies' (Dror, 1989: xiii) may be found in Greek and Renaissance political theory, but it is not so evident in Roman civilization. Policy in the sense of choosing between options is, he suggests, an idea which may also be discerned in mercantilist writings on trade. Heidenheirner (1986), in a study of the policy concept in Europe, records that from the late eighteenth century onwards, professorial chairs in 'Polizey' science were established in an effort to systematize knowledge about administration and social welfare. The word's Anglo-Saxon notion of 'policy' does not travel well. In many European languages there are PUBLIC POLICY problems over distinctions to be made between 'policy' and 'politics' As Ostrom and Sabetti observe: 'Policy as used in English is not easily rendered into French, German, Italian and Spanish.' Lerner and Lasswell's The Policy Sciences in the United States (1951) had to be I translated into French as Les Sciences de la politique aux Etats-Unis (Ostrom and Sabetti, 1975: 41). The modern meaning of the English notion 'policy' is that of a course of action or plan, a set of political purposes - as opposed to 'administration' (Wilson, 1887). Above all, the modern meaning of the word, dating from the post-Second World War period in particular, is that of policy as a rationale, a manifestation of considered judgement. Imagine, for example, politicians admitting that they do not have a policy on xl A policy is an attempt to define and structure a rational basis for action or inaction. As the state changes its mode of legitimating discourse, so the func- £ tion of 'policy' has altered. The modern liberal democratic state, post-Second World War, was to be a system which sought to define its I legitimacy in terms of policy. Hogwood and Gunn (1984:13-19) specify ten uses of the term 'policy' in this modern sense: as a label for a field of activity; as an expression of general purpose or desired state of affairs; as specific proposals; as decisions of government; as formal authorization; as a programme; as output; as outcome; as a theory or model; as a process. The meaning of the word policy must also be understood in a more historical context. For, like the concept of public, the changing meaning of policy tells us much about the change in policy in practice. In English, 'policy' has a rich and complex meaning. In Shakespeare, for example, we may find four distinct uses: prudence, a form of government, affairs and administration, and as 'Machiavellianism' (see below: 1.6). Policy encompassed the arts of political illusion and duplicity. Show, outward appearance and illusions were the stuff of which power was made. Shakespeare employed the terms of Machiavellian philosophy that were well known at the time. Power cannot be sustained purely with force. It needs, in a Machiavellian sense, policy: and 'policy sits above conscience', as the bard tells us in Timon of Athens. META ANALYSIS 15 ♦ Policy as craftiness It is perhaps in the work of Shakespeare's great contemporary, Marlowe, that we may see one of the most interesting illustrations of the use of a Machiavellian notion of policy. In Marlowe's nlav The Jew of Malta the notion of policy has a central role; the word itself appears many times in the text. A knight, for example, refers to it as 'simple policy', to which Barabas adds later, 'Ay, nolicy that's their profession, And not simplicity as they suggest.' Policy has a duality of meaning: simple and scheming. Policy involves creating a plausible story which secures the purposes of the plotter: policy is acting a part. And, as Ithamore says, The meaning has a meaning.' By his policy Barabas gets 'no simple place' as the Governor of Malta. He plays off Turk against Christian and thus makes a 'profit' from his policy. Fiction is often stranger than fact In Yes Prime Minster, by Jonathan Lynn and Antony Jay (1987), we find an interesting illustration of the way in which a policy may be little more than elaborate window-dressing. Sir Humphrey, discussing a plan for reducing unemployment, observes that in reality the PM is 'only trying to look as if he is trying to reduce unemployment. This is because he is worried that it does not look as if he is trying to look as if he is trying to reduce unemployment' (p. 26). ♦ As we shall see (1.6) Francis Bacon, a contemporary of Shakespeare and Marlowe, also defined policy in terms of rational cunning. However, over time this notion of policy as politics and of politics as policy is replaced by the idea of policy as political whilst carrying it out, or implementation as 'administration' or 'bureaucracy'. With the development of industrial society in national states and its consequent administrative forms, bureaucracy, as Max Weber demonstrated, became the expression of the rational component of the state, whose function it was to carry out the will of its political - elected -masters. Bureaucracy derived its legitimacy from its claims to being non-political, whilst politicians claimed that their authority rested on the approval of their policies or 'platforms' by electorates. Policy therefore as a term becomes an expression of political rationality. To have a policy is to have rational reasons or arguments which contain both a claim to an understanding of a problem and a solution. It puts forward what is and what ought to be done. A policy offers a kind of theory upon which a claim for legitimacy is made. With the development of modern electoral and party systems in industrial societies policy discourse became the main mode through which electorates engaged with 'politics' and rival political elites. The politician is expected to have 'policies' as a shop is expected to have goods to sell. In 16 PUBLIC POLICY the Schumpeterian 'realistic' sense of democracy, the 'policy' or 'plank' was the essential currency of democratic exchange: the democratic method is that institutional arrangement for arriving at political decisions in which individuals acquire power to decide by means of a competitive struggle for the people's vote ... all parties will of course, at any given time, provide themselves with a stock of principles or planks and these principles or planks may be the characteristic of the party that adopts them and as important for its success as the brands of goods a department store sells are characteristic of it and important for its success ... Party and machine politicians are simply the response to the fact that the electoral mass is incapable of action other than a stampede ... The psycho-technics of party management and party advertising, slogans and marching tunes, are not accessories. They are the essence of politics. (Schumpeter, 1974: 269-83) (On the role of party management and advertising, see Franklin, 1994, and section 2.3.2 in this volume.) The idea of policy as a 'product' or 'plank' consequently developed a neutral connotation, far removed from the Machiavellian sense displayed in Shakespeare or Marlowe. Policy and politics now (in English at least) become quite distinct terms. The language and rhetoric of 'policy' thus became the main instrument of political rationality. As Lasswell observed: The word 'policy' is commonly used to designate the most important choices made either in organized or in private life ... 'policy' is free of many of the undesirable connotations clustered about the word political, which is often believed to imply 'partisanship' or 'corruption'. (Lasswell, 1951b: 5) In liberal democratic systems, political elites have to give rational reasons for what they propose or what they have done. Peter the Great had only to say that he did not like beards for beards to be shaved off. In regimes which operate under a code of religious beliefs, it may well be enough that an edict or decision is justified in accordance with a religious precept for it to be considered legitimate. In societies that are not so informed by religious values, politicians as policy-makers have to claim that they are doing something after rational consideration of the facts: in other words, we expect governments to have 'a policy'. 1.3 The development of the policy approach The development of policy analysis must be placed in the context of this rationalization of the state and politics as a 'policy-making' activity. The emergence of the methods of natural science provided the META ANALYSIS 17 essential framework within which the study of society and 'public administration' was to grow in the nineteenth and twentieth centuries. As Trudi Miller notes: Natural science represents the approach to public administration and political science that has prevailed for most of the twentieth century ... The implicit assumptions of this nature-focused approach are that (1) the laws that govern human behaviour exist independent of human control, and (2) the units of analysis in social systems are highly similar over time and space. The conventional quantitative methods of social science reflect these implicit empirical assumptions. (Miller, 1984: 253) As we shall see later (4.5) this was to prove especially true in the case of analysis deployed to measure/evaluate the 'performance' of organizations, people and policies in an 'assumptive world' (Young, 1977, 1979: see 3.7.7) of 'reliable relationships and measurable phenomena' (Kirlrn, 1984:161). The Enlightenment notion that the world was full of puzzles and problems which, through the application of human reason and knowledge, could be 'solved' forms the background to the growth of the policy approach. What Newton had done to the laws of planetary motion became a model for what it was possible to do with knowledge of human society. Thus we may chart the development of the policy sciences in terms of the desire for knowledgeable governance, that is, the acquisition of facts and 'knowledge' about 'problems' so as to formulate 'better solutions'. As Max Weber showed, the growth of industrial civilization brought about a search for more rational forms of organization for the state, commerce and industry (Weber, 1991:196-252). Out of this was to emerge the kind of separation of policy-making as a political function from administration as a bureaucratic function. At the same time, the desire for a more rational approach to social and other 'problems' was manifested in the growth in the capacity of the state to acquire and store information, and in the development of empirical research such as social surveys (see 2.2.1). Later, in the early twentieth century, this idea that government could, by making policies 'solve' problems, not least those of 'the economy', meant that the social sciences began to establish a new relationship to politics and government. In the 1930s social scientists, the most famous of whom was the economist John Maynard Keynes, could claim that, if government was to have any chance of dealing with the problems of the day, it had to recognize the need for a more informed, theoretically driven approach to governing. In the future, he predicted, it would be the ideas of economists rather than political interests which would shape decision-making (Keynes, 1936). 18 PUBLIC POLICY Other 'policy'-focused forms of inquiry were to develop in the period between the wars and up to the 1950s and early 1960s. For the greater part of the period fields such as sociology, psychology, political science, social administration, management, and natural sciences with policy implications remained related, but in no sense formed a common approach. It was in America where moves towards a more unified approach to the study of public problems and policy really began in the work of Harold Lasswell (see, for example, Lasswell, 1930a, 1948b), whicn cuiiiinuueu in me puDiicauon or Lerner and Lasswell's volume on the 'policy sciences' (1951). ♦ Lasswell and the idea of policy sciences and the role of the policy analyst Sciences are policy sciences when they clarify the process of policy-making in society, or supply data needed for the making of rational judgments on policy questions ... If we get rid of the standoffishness that has kept men of knowledge apart in our civilization, we can more conveniently come together in research teams that are capable of contributing the knowledge needed by the democratic polity ... Today we are living in a world of ever-deepening shadow, in which basic democratic values are challenged as never before and in which even the survival of the human species is at stake. Under these circumstances it makes sense to develop a strategy of using our limited intellectual resources for the defense and extension of our values. The term 'policy' is used to indicate the need for clarifying the social ends to be served by a given allocation (including self-allocation) of scientific energy. {Lasswell, 1948b: 122) The policy sciences includes (1) the methods by which the policy process is investigated, (2) the results of the study of policy, and (3) the findings of the disciplines making the most important contribution to the intelligence needs of our time. (Lasswell 1951b: 4) We can think of the policy sciences as the disciplines concerned with explaining the policymaking and policy-executing process, and with locating data and providing interpretations which are relevant to the policy problems of a given period. The policy approach does not imply that energy is to be dissipated on a miscellany of merely topical issues, but rather that the fundamental and often neglected problems which arise in the adjustment of man in society are to be dealt with. The policy approach does not mean that the scientist abandons objectivity in gathering or interpreting data, or ceases to perfect his tools of inquiry. The policy emphasis calls for the choice of problems which will contribute to the goal values of the scientist, and the use of scrupulous objectivity and maximum technical ingenuity in executing the projects undertaken. The policy frame of reference makes it necessary to take into account the entire context of significant events (past, present, and prospective) in which the scientist is living ... It is probable that the policy science orientation ... will be directed toward improving the knowledge needed to improve the practice of democracy. In a word, the special emphasis is upon the policy sciences of democracy, in which the ultimate goal is the realization of human dignity in theory and practice. (Lasswell, 1951b: 14-15). ♦ META ANALYSIS 19 »♦♦ Harold D. Lasswell, 'The Emerging Conception of the Policy Sciences', 1970 Lasswell's article set out the state of the policy sciences in 1970 in the first number of the iournal Policy Sciences. He introduced the idea of knowledge of the policy process and in the nolicy process, and argues that the distinctive outlook of the policy science is that it is problem-orientated. This problem focus means that the subject aims to be multidisciplinary and involved in the synthesis of ideas and techniques. What the policy scientist brings to the analysis of problems is a creativity in which there is a 'creative rearrangement', and an enlargement of the conceptual map which defines the problem as perceived by specialists: The contemporary policy scientist perceives himself as an integrator of knowledge and action, hence as a specialist in eliciting and giving effect to all the rationality of which individuals and groups are capable at any given time. He is a mediator between those who specialize in specific areas of knowledge and those who make the commitments in public and private life ... Both the intellectual community and the community at large are beginning to acknowledge the indispensable place of the integrator, mediator, and go-between ... (Lasswell, 1970a: 13-14) The policy sciences were, therefore • contextual; • multi-method; • problem-orientated. A major feature of this orientation towards policy as a knowledge process was the designation of stages and functions within policy-making (see 1.10). Policy analysis as public therapy We cannot fear the policy scientist because, unlike the politician, the policy scientist is not a driven personality. As Lasswell discusses him we get the impression that sooner or later political conflicts must yield to the policy scientist's implacable logic and empirical data, just as in the therapeutic situation, the patient's personal conflicts gradually yield to the analyst's expert ministrations ... We cannot escape the parallel here between the function of policy science for the political scientists and the function of the training for the fledgling psychotherapist. The training analysis arms the young therapist with insight into his or her own deepest motives, thus preventing the projection of the therapist's conflicts onto patients and the consequent perpetuation rather than curing of illness. Similarly, learning to be a policy scientist is self-therapy, for it obliterates the social scientist's lust for power. Thus, policy science is therapeutic and pragmatic: the social physician heals himself while learning to heal the polity. (Merelman, 1981: 496) ♦ The policy sciences soon settled into two main approaches which, as Lasswell suggested, could be defined in terms of knowledge in the policy process and knowledge of the policy process (Lasswell, 1970a): 20 PUBLIC POLICY • 'policy analysis': concerned with knowledge in and for the policy process; • the analysis of the policy process: concerned with knowledge about the formation and implementation of public policy. Policy analysis may be traced back to the war years, in particular to the introduction of OR (operations research), and techniques of economic analysis. Among the first kinds of policy analysis, therefore, was that which took place in economic policy-making and defence However, later in the 1960s in the US and elsewhere governments increasingly required more information and analysis about education, transport, urban planning, health, and so on. In America this expansion of government as a problem-solver became associated with the Kennedy-Johnson 'New Frontier' and 'Great Society' programmes. These programmes called forth a new kind of applied methods of investigation whose primary goal was to analyse 'problems' and to develop options or alternatives which could ameliorate or solve them. Meltsner (1976: 2) notes that the modern terms 'policy analysis' and 'policy analyst' began to be used frequently in the 1960s by Dror (1967), in various government papers (US Congress, 1969) and, Meltsner believes, was probably first used by Charles Lindblom, in 1958. This kind of policy analysis (what Wildavsky (1979) terms 'speaking the truth to power') was in all essentials a belief in social science as a form of engineering or medicine. Knowledge of society could provide a way of making it better. Furthermore, this growth of analysis for government was allied to the export from the defence sector and the business world ideas about how decision-making as a process could be made more effective. One major source of these ideas was operational or operations research, a method of analysis which had been developed in Britain and America during the Second World War. The other main influence over policy analysis in government were theories and techniques borrowed from management in the private and corporate sector. As we shall see later, these two influences were to come together via the personification of 1960s' analysis: Robert McNamara, President Kennedy's Defense Secretary. The approach to policy analysis which emerged in these years was in a relatively short time to percolate through to other policy sectors and departments in the US and elsewhere. By the 1960s, more and more was going into the black box of the policy-making 'system'. Liberal-democratic governments were increasingly being called upon to take responsibility for a growing range of social, economic and other problems. In the USA it is a period closely identified with the Kennedy-Johnson years. Typical of the literature which was to emerge in the 1960s was the volume edited by Bauer META ANALYSIS 21 and Gergen, The Study of Policy Formation (1968), which contained chapters reviewing the then state of the art. The book contained an approving foreword from Robert C. Wood, then Under-secretary at the Department of Housing and Urban Development. Wood commends the book for the way in which it assists practitioners such as himself in the task which President Johnson expressed as 'knowing as well as doing what is right'. Policy analysis therefore evolved in an era in which government was seen as a 'problem-solver' and the political system as a problem-processor. As Keith Hope observed of American politics at this time: The typical American word for an unsatisfactory social state is 'problem', something, that is, which can be solved and thereby disposed of; and the typical word for ameliorative social action is 'program', something, that is, which has a pre-ordained beginning, middle and end. Thus it is that social scientists in the United States frequently talk as if it were possible to jump to the goal of the 'Great Society', without first passing though a period of a welfare state ... (Hope, quoted in Sharpe, 1975:16) The other kind of analysis - which developed pari passu with the problem focus in the 1960s - was the analysis of the policy process as an alternative focus to the shidy of constitutions, legislatures and interest groups, and public administration. This policy focus in political science is most closely associated with the contribution of four people: Harold Lasswell, Herbert Simon, Charles Lindblom and David Easton. Their ideas will feature prominently throughout this book, and there is no better starting point for the study of policy-making and the role of policy analysis than to read their early works and follow the development of their thought. Lasswell, perhaps, stands out as the pre-eminent moving spirit behind the growth of a policy approach. His writings on public policy may be dated back to the 1930s, when he was inspired by the Chicago School to be concerned with problems and to take a multidisciplinary approach, hi the 1940s, for example, he was instrumental in setting up an early 'think-tank', the American Policy Commission, whose aim was to 'close the gap between knowledge and policy' by fostering a constructive dialogue between social scientists, businessmen, and policy-makers (J.A. Smith, 1991: 105). He was also closely involved with one of the most important 'think-tanks', the RAND corporation (as chairman of the board of trustees of the RAND graduate school; on RAND, see 2.8.2). Lasswell (1956) was one of the early attempts to formulate a set of 'stages' in the policy process (see 1.10). Herbert Simon's contribution to the development of the policy approach has been without doubt more far-reaching than any other 22 PUBLIC POLICY single theorist. Given the multidisciplinary nature of public policy, the fact that Simon's work has impacted on a range of social sciences -including economics, psychology, management, computer science, sociology and political science - means that wherever the student turns, there he is. His concern with human decision-making has centred on the idea of rationality as 'bounded' but capable of improvement. This theme he has explored both theoretically and experimentally. Simon's idea of examining decision-making in terms of a sequence of rational stages: intelligence, design, and choice, has formed a central element of policy analysis. Simon will figure prominently throughout this book, particularly in 3.4.2 and 3.7.4. The third key contributor to the development of policy analysis as concerned with the 'process' of making policy is Charles Lindblom, who is best known for his advocacy of an alternative to Simon's rational approach in the form of 'incrementalism'. His article on the 'science of muddling through', published in 1959, is a classic text in the literature of policy studies. It remains as perhaps the single most important contribution to the formation of a theory of the policymaking process. Over the years Lindblom's thought has evolved beyond his original argument - some might argue to the point where there are two different kinds of Lindblom. We shall examine his ideas more fully in Parts Two and Three. In criticizing the 'rational' model as expounded by Simon and others, Lindblom also rejected the idea that thinking in terms of 'stages' or 'functional' relationships (see Easton, below) was of any real value to the study of the policy process. The models which owe their inspiration to Lasswell, Simon and Easton were viewed by Lindblom as thoroughly misleading. Instead, Lindblom (1968) proposed a model which took account of power and interaction between phases and stages. As he explains in the most recent edition of his textbook on the policy process: 'Deliberate, orderly steps ... are not an accurate portrayal of how the policy process actually works. Policy-making is, instead, a complexly inter-active process without beginning or end' (Lindblom and Woodhouse, 1993: 11). Lindblom suggests that in studying the policy process we should take account of elections, bureaucracies, parties and politicians, and interest groups, but also of 'deeper forces' - business, inequality, and the limited capacities of analysis - which structure and distort the policy process. ♦ Lindblom's framework C.E. Lindblom and E.J. Woodhouse, The Policy-making Process, 3rd edn, 1993 META ANALYSIS 23 What are the limits of analysis in the policy process? What is the role of analysis in a democracy? Conventional government and politics and policy: the imprecision of voting; the impact of elected functionaries; bureaucracy and policy-making; interest groups and policy-making. The broader influences: the role of business; political inequality; impaired inquiry. How can policy-making be improved (given the above)? ♦ Figure 1.1 The Eostonian 'black box' model c (D E > c O) CD sz F The intra- societal environment The extra-societal environment The flow of effects from the environments The political system Information feedback Conversion JUUXTLrLTV of demands into outputs Information feedback A U T H O R I T I E S Outputs Feedback loop- Notes: The intra-societal environment: • ecological system • biological system • personality system • social system Source: Adapted from Easton (1965:110) The extra-societal environment: • international political systems • international ecological systems • international social systems Finally, the work of David Easton (1953,1965), although not regarded as primarily 'public policy', has made as vital a contribution to the establishment of a policy approach as the other scholars we have mentioned, in that it provided a model of the political 'system' which greatly influenced the way in which the emerging study of policy PUBLIC POLICY (outputs) in the 1960s began to conceptualize the relationship between policy-making, policy outputs and its wider 'environment' (see Figure 1.2). The main characteristics of the Eastonian model is that of viewing the policy process in terms of received inputs, in the form of flows from the environment, mediated through input channels (parties, media, interest groups); demands within the political system (withinputs) and their conversion into policy outputs and outcomes (see figure 1.3). The frameworks which have dominated the field from the 1960s onwards derive from the combination of the 'stages' approach set out by Lasswell, Simon and Easton's 'political system' model. The textbooks which, in Kuhn's sense (see 1.8), provided the 'normal science' of policy analysis were, for the most part, derived from the fusion of Lasswell, Simon and Easton's models of decision-making and the political 'system'. (See, for example, Jones, 1970; Frohock, 1979; Dye, 1972; Sharkansky, 1970.) Figure 1.3 The policy process as inputs and outputs Input------------------------► Policy--------------- -------► Output-----------*--------- i Perception/identification Regulation Application Organization Distribution Enforcement Demand Redistribution Interpretation Support Capitalization Evaluation Apathy Ethical ruling Legitimation Modification/adjustment Retreat/disavowel Sources: Frohock (1979); Jones (1970) The combination of rational stages and systems approaches thus afforded a more dynamic model of policy-making, and a basis for understanding policy in terms other than institutional and constitutional arrangements; although from Lindblom's point of view the models which developed have served more to obscure than to illuminate the policy process. Along with Easton's a number of other structural functional or 'systems' models had an important role in the development of new models of the policy process in the 1960s. Chief amongst them were those of G.A. Almond and Karl Deutsch (on Deutsch, see 3.7.6). The most comprehensive introduction to Almond is Almond and Powell (1966 and later editions) and Almond et al. (1993). The latter sets out a model of the political system as composed of inputs (interest articulation), process functions (interest aggregation, policy-making, policy implementation and adjudication), and policy function META ANALYSIS 25 (extraction, regulation and distribution). Policy output is fed back into the political system, which is located in a domestic and international environment. This (1993) version acknowledges that it has sought to take more account of the role of institutions than in the past, when political scientists were hell bent on ignoring the fact that institutions, rules and constitutions do actually matter (see, for example, March and Olsen (1989); and 2.11.6, 3.6) The theorists who had most influence on the development of models for analysing the policy-making process were American, but there were a few exceptions to the dominance of American policy science. A British theorist - and practitioner - whose work was important but had far less influence on the way in which the policy approach evolved was Sir Geoffrey Vickers. We shall examine his ideas in 3.7.5. Vickers's work stressed the importance of analysing the interaction of value judgements and reality judgements and was a major synthesis of psychological, cybernetic and political ideas, as well as an extensive administrative experience. It is to be hoped that with the rise in interest in the role of values in the policy process his work will find a new relevance for the study of public policy in the 1990s. ♦ Sir (Charles) Geoffrey Vickers (1894-1982) Born in Nottingham, he went up to Merton College, Oxford in 1913 to read classics. During the war he was awarded the V.C. and the croix du guerre. On leaving the army he returned to Oxford to complete his degree and qualified as a solicitor. During the Second World War he was put in charge of economic intelligence in the Ministry of Economic Warfare. He joined the new National Coal Board and became Board member in charge of personnel and training in 1948 and retired in 1955. Vickers became involved in voluntary work in support of medical research especially in psychiatric and mental illness. His retirement gave him the opportunity to pursue his ideas, resulting in ten books and many papers, articles and lectures. His work became widely known and respected in the US and Canada. Nevil Johnson records that: The problem which chiefly preoccupied Vickers was how individuals can best fulfil the requirements of social cooperation in conditions of accelerating economic and scientific change. He came to reject moral and economic individualism and argued that institutions are necessary conditions of satisfactory social coexistence. Influenced by Michael Polanyi he saw the achievement of an adequate understanding of institutions as an epistemological challenge: individuals have to grasp how their actions always involve the regulation of relationships with others, and this occurs only through the exercise of judgement. Consequently much of his work is devoted to the analysis of judgement in terms of what he called appreciative behaviour'. Though appreciation and judgement express individual capacities, 'ickers never saw the individual as isolated or sovereign, but rather as defined by the ilationships he has. He believed that social institutions are best analysed in terms of items and his published work ... made far-reaching contributions to systems thinking in its